Sei sulla pagina 1di 43

Investment Preference, Risk Perception, and Portfolio Choices under Different Socio-Economic Status: Some Experimental Evidences from

Individual Investors

Sheng-Hung Chen Assistant Professor Department of Finance Nan Hua University, Chiayi, Taiwan Mail Address: 32, Chung Kung Li, Dalin, Chiayi, 62248, Taiwan Phone: +886 5 2721001 ext 56541 Mobile: +886 9 35575138; Fax: +886 5 2427172 E-mail: shenghong@mail.nhu.edu.tw

Chun-Hung Tsai Graduate Student Department of Finance Nan Hua University, Chiayi, Taiwan Mail Address: 32, Chung Kung Li, Dalin, Chiayi, 62248, Taiwan E-mail: n1247228@yahoo.com.tw

January 2010

Investment Preference, Risk Perception, and Portfolio Choices under Different Socio-Economic Status: Some Experimental Evidences from Individual Investors

Sheng-Hung Chen 1,* Chun-Hung Tsai 1


1

Department of Finance, Nan Hua University, Chiayi, Taiwan

Abstract

Identifying key factors influencing individual investors decision to make portfolio choices is of importance to understand their heterogeneous investment behavior, however previous research based on experimental evidence is extremely space especially in terms of perspective from individuals perceptions on risk tolerance and risk perception. This paper explores individual investors preference for portfolio choices and empirically investigates impacts of risk tolerance and risk perception on their investment decision. Specifically we decompose socio-economic stats difference in investment preference for portfolio choices with respect to investors gender, age, income level, marital status, education level and household income level. Using conjoint analysis on investment experiments to obtain some evidences from a sample of 285 respondents in survey, our results indicate that investors decisions to make their portfolio choices are significantly and negatively related to personal labor income level, female and married investors, and square of overall scores on risk perception while however the case is positively associated with their household income level and the level of risk tolerance. This finding implicates that investor with higher risk tolerance level shows higher likelihood to make their investment decision on portfolio choices. Moreover, individual investors would like to make the investment decision on portfolio choices with higher weights of bond investment and saving. Interestingly, they do not prefer to make portfolio choices with international portfolios and short-run investment horizon. Besides, it is found that male investor demonstrates much preference on portfolio choices with higher percentage of total return. Keywords: Portfolio Choice, Investment Preference, Individual Investors, Behavioral Finance, Risk Perception, Risk Tolerance, Socio-Economic Status, Conjoint Analysis, Conditional Logit Model, Survey Data

* Corresponding author Sheng-Hung Chen, Assistant Professor Department of Finance, Nan Hua University, Chiayi, Taiwan Mail Address: 32, Chung Kung Li, Dalin, Chiayi, 62248, Taiwan Phone: +886 5 2721001 ext 56541; Fax: +886 5 2427172; Mobile: +886 9 35575138 E-mail: shenghong@mail.nhu.edu.tw

1. Introduction Traditional mean-variance theory has been widely investigated by both practitioners and financial economists. The significant characteristics on modern investment analysis are simple to measure the risk and return trade-off in terms of the mean and variance of investment. Some investment applications related to such issues are portfolio analysis, asset pricing, and performance measurement. However, there are other important features of typical portfolio choices that are not comprehensive described explicit in terms of individual investors wealth level, investment information, and investment horizon. Portfolio choice for individual investor is mainly determined by both investment attributes and investors objective judgment. Specifically, investors investment decisions on various financial products available to choice are extremely related to different risk measures where most investors usually rely on their risk perceptions and risk tolerances to make portfolio choices, but current research concerning the behavior finance perspective from the impact of risk perception and tolerance on portfolio choices of individual investor is surprisingly scarce. This issue is very important to know how individual investors do their heterogeneous investment decisions to make different portfolio choices. This paper is aimed to investigate empirically the potential impacts of risk perception and risk tolerance on portfolio choice for individual investor under risk uncertainty in terms of socio-economic status differentials. Therefore, evaluating the influence of risk perception on investment decision is helpful for more understanding about investment intention and attitude for individual investor. In particular, the socio-economic status differentials would play an important role to investigate the impact of risk perception on portfolio choice for individual investor, but this issue is yet addressed in previous studies. Moreover, the asset management institution is interesting to understand how individuals allocate their asset. Most of financial planning worker or financial adviser suggests that investors alter their asset allocation between different portfolio choices to reflect risk tolerance of individual investors. In the context of investment advice for individual investors, financial institutions have used so-called risk profiles for their clients. Risk profiles used by different banks in different countries all have in common that they investigate both the time horizon of the investors and on their risk perceptions (Veld and Veld-Merkoulova, 2008). Portfolio choice is the core investment decision for individual investor who tries to avoid money loss and to gain return on their investment plan (Veld and Veld-Merkoulova, 2008; Frijns et al., 2008). Even though both economists and psychologists have studied
1

decision making under risk uncertainty, however, much research relates to simple choices. Risk perceptions for individual investors are explored by experimental design on their potential concepts about risk exposures. Some authors indicate that investment decision is determined by financial risk taking (Grable et al., 1999) and risk tolerance (Veld and Veld-Merkoulova, 2008). Hence, identifying key factors influencing portfolio choice for individual investors is of importance, particular in the role of risk perception (Grable et al., 1999; Hallahan et al., 2004). The main contribution of this paper is to recognize the relationship among portfolio choice, socio-economic factors, risk perceptions and risk tolerance and asset management institution for individual investor by measuring the risk perception and risk tolerance of individual investor for portfolio choice. Financial risk tolerance, reflecting a persons attitude towards taking on risk, is a complex psychological concept. However, because of the complexity of the attitudinal construct, a sophisticated psychological testing instrument is required to elucidate a persons attitude to financial risk. Psychometrics is that area of psychology dealing with the design and analysis of measurements of human characteristics. Financial risk tolerance is a term widely used in the personal financial planning industry to refer to an investors attitude towards risk. It can be defined as the amount of uncertainty or making a financial decision. Whatever measured for the purpose of self-assessment or for documentation of investment suitability, financial risk tolerance is assumed to be a fundamental issue underlying a number of financial decisions. Because of those reasons, researchers have been interested in understanding why the internal factors and externals factors will interact and influence individual investors invest decision. Moreover, it is noteworthy that socio-economics status does play an important and differential role on risk-taking and risk-tolerance based on previous research, indicating that factors such as gender (Hallahan et al., 2003; Hallahan et al., 2004; Veld and Veld-Merkoulova, 2008), age (Bakshi and Chen, 1994; Brown, 1990; Dahlback, 1991; Mclnish, 1982; Morin and Suarez, 1983; Palsson, 1996; Veld and Veld-Merkoulova, 2008), marital status has also been postulated to impact on financial risk tolerance; however, the exact nature of the relationship is not clear. One view asserts that single people are more risk tolerant than married individuals because they have less responsibilities than married people, particularly in respect to dependents, and face less social risk (that is, potential loss of esteem) when undertaking risky investments (Roszkowski et al., 1993, Grable, 2000; Frijns et al., 2008), occupation, income (Grable, 2000; Hallahan et al., 2003; Hallahan et al., 2004; Veld and Veld-Merkoulova, 2008), may influence a individuals level of risk taking and risk-tolerance in portfolio choice. Because of each
2

individual investor might be influenced by psychology factor for their investment decision (Veld and Veld-Merkoulova, 2008). Similarly, research findings in relation to income also uniformly support a positive relationship. That is, higher levels of income and wealth have been found to be associated with higher levels of risk tolerance (Friedman, 1974; Cohn et al., 1975; Blume, 1978; Riley and Chow, 1992; Grable and Lytton, 1999; Schooley and Worden, 1996; Shaw, 1996). Therefore, socio-economic status differential is noteworthy to explore the investment decision on portfolio choice. It is found that an increase in the risk-return trade-off (more return for a given amount of risk) increases demand for the risky assets, an increase in the saving accounting rate has a positive impact on the demand for the riskless asset, and the interaction of investors risk aversion with the risk-return trade-off decreases their preference for the risky asset. There are two methodology used in this study as survey and conjoint analysis. We design survey to collect primary information about investment behavior on portfolio choice from individual investors. Our survey is composed of four sections. First section is to measure individual investors degree of risk tolerance and risk perceptions. Second section aims to understand investors portfolio choice, according to their demographic status background, risk tolerances, risk perceptions and investment preference. Third section is the behavior of investors, they based upon their portfolio choice (or products choice). The last category is investors preference about asset management institution attributes. Conjoint analysis is applied to examine how individuals derive preferences for products or services by assuming that they sum individual utilities for each attribute to obtain the overall utility for financial assets or portfolio choices. In this paper of our study design structure, we separate four parts in my thesis structure. Chapter 1, introduction is going to briefly explanation about our research from motivations to summary of research conclusions. Chapter 2 is literature review, in order to find evidences to support or to prove our research purpose and expected results of our study from previous studies related. Chapter 3 is the research design is including the design of questionnaire and describes the sample collection that we collected. Chapter 4 are empirical results principal to analysis the relationship among socio-economics status, risk tolerance, risk perceptions and portfolio choice of individual investors and illustrate may bring some influence individual investor on their portfolio choice. The Chapter 5 are conclusions of this paper is trying to collation all the summary of our empirical results and describe what kind of useful information that we can provide from our research.

2. Literature Review In this section, the literature review including three parts. First, behavior finance perspective of individual investor. Second, individual investors risk perception, risk tolerance and portfolio choice. Third, individual investors socio-economic status differential and risk tolerance. The results for gender, education level and income level are consistent with the earlier literature. Previous literature indicating those factors on risk-taking and risk tolerance are gender, age, marital status, occupation, income level, education level and economic environments expectations, which might influence an individual investors level of risk taking, but the factor of education level might not. Those studies are classified by three catalogers. The first one focuses on behavior finance perspective of individual investors in their products choice. The second one is to evaluate the impact of risk perceptions and risk tolerance for individual investor by risk measure tools selected, risk perceptions respond, risk tolerances questions selection and various portfolio choice. The third one is to explore and realize the relationship between socio-economic status variables and risk tolerance, the linkages are including all of background information about individual investors risk related opinion and they how to reflected those risk related perceptions on their portfolio choices and risk measure tools. 2.1 Behavior Finance Perspective of Individual Investor As a result of traditional finance theory appears to play a limited role in understanding this issues such as (1) why do individual investors trade, (2) how do they perform the task, (3) how do they choose their portfolios to conform their conditions, and (4) why do returns vary so quickly even across stocks for reasons other than risk. In the new arena of behavior finance or so-called behavior economic, we could to interpret about individual investors behave in their invest choice more completely. Most of behavioral finance researchers often claimed that the reality results presents no unified theory unlike traditional finance theory appears expected utility maximizations using rational beliefs. Its means those scholars in this field actually postulate whole investors in financial market are rationales; they cant influenced through any factors only maximum profit for themselves. Most authors show behavior finance perspective on individual investor, such as Deaux and Emswiller (1974), Lenney (1977), Maital et al. (1986), Thaler and Johnson (1990) and
4

Beyer and Bowden (1997). Those authors are to exclaim that individual investor would demonstrate different risk attitude when facing investment alternatives. Later instruction in our research, we called risk perception and risk tolerance of individual investor. Comparing with previously research, current study is to focus on external factors and psychological factors how to affect investors investment decision and portfolio choice. For instance, Annaert et al. (2005), Wang et al. (2006) indicate the impact of information asymmetric problem on investor behave, this is another subject in behavioral finance field. Most of these researches are pay close attention to behavioral finance, especially in financial products choices (investment) and behave of individual investor invest related. 2.2 Risk Perception, Risk Tolerance and Portfolio Choice Roszkowski (1998) noted that calculating individual investors level of risk tolerance is a difficult process because risk tolerance is an elusive, ambiguous concept. Many authors have suggested that risk taking is constant across situations, but evidence indicates that, for instance, a individual investors level of risk tolerance for physical activities is not a suit gauge of risk taking in financial situations (Roszkowski, 1998; Rowland, 1996). Financial risk tolerance is defined as the maximum amount of uncertainty that someone is willing to accept when making a financial decision. Although the importance of assessing financial risk tolerance is well documented, in practice the assessment process tends to be very difficult due to the subjective nature of risk taking (the risk of investor willing to reveal their risk tolerance) and objective factors such as Grable and Joo (1997), Grable and Lytton (1999), and Grable (2000). Risk tolerance represents one persons attitude towards taking risk. This indicated is an important concept that has implications for both financial service providers (asset management institution or other financial planner) and consumers (investors). For the latter, risk tolerance is one factor which may determine the appropriate composition of many assets in a portfolio which is optimal and satisfied investors invest preference in terms of risk and return relative to the needs of the individual investors Droms, (1987), Hallahan et al., (2004). There are some empirical evidence showing the impact of risk perception; risk tolerance and socio-economic on portfolio choice, for instance, Carducci and Wong (1998), Grable and Joo (1997), Grable and Lytton (1999), Grable (2000), Hallahan et al., (2003), Hallahan et al., (2004), Frijns et al., (2008), and Veld and Veld-Merkoulova (2008). In terms of different risk perception or risk tolerance level, individual investor may show different
5

reaction base upon their psychology factor and economic situation, which would lead to heterogeneous portfolio choice for individual investors. For this reason, it is crucial to recognize and attitudinal how individual investors with different risk perceptions and risk tolerance make their invest products choice on investment plan, in particular socio-economic status differentials may make their choice vary and difference. 2.3 Investors Socio-Economic Status and Risk Tolerance Some researchers have indicated that the validity of widely used demographics as determinants of risk tolerance is noteworthy as the relationship between socio-economic status differences including gender, age, income level, net assets, marital status, educational level and investment decision or portfolio choice. With regard to the financial risk tolerance literatures, there is much interest in the demographic determinants and risk attention (involving three risk types: risk aversion, risk moderate and risk seeking) is particularly focused on age, gender, education level, income level, marital status, the number of dependents and net assets. Specifically, although debate remains on some issues, a range of common findings are generally observed. There are five phenomenons in socio-economic status variables differential and portfolio choice as the following: First, risk tolerance decreases with age (e.g., Morin and Suarez 1983; Roszkowski, Snelbecker, and Leimberg 1993). Second, females have a lower preference for risk than males (e.g., Roszkowski, Snelbecker, and Leimberg 1993; Grable 2000). Third, risk tolerance increases with education level (e.g., Roszkowski, Snelbecker, and Leimberg 1993; Haliassos and Bertaut 1995). Fourth, risk tolerance increases with income level and net assets (e.g., Cohn et al. 1975; Roszkowski, Snelbecker, and Leimberg 1993; Bernheim, Skinner, and Weinberg 2001). Fifth, single (i.e., unmarried) investors are more risk tolerant than married (e.g., Roszkowski, Snelbecker, and Leimberg 1993). 2.3.1 Age In the early 1960s Wallach and Kogan (1961) began studying relation ships between risk tolerance and age. Many studies have shown that age interact with financial information and issues differently. To survey the literature, this indicates that younger have different attitudes toward financial decisions than elder (Wallach and Kogan, 1961; McInish, 1982; Morin and Suarez, 1983; Palsson, 1996). The majority of the published research studies
6

examining the relationship between age and risk tolerance have found that risk tolerance decreases with age. They found that older individuals tended to be less risk tolerant than younger persons. This finding is widely held to be true, both among practitioners and researchers (Bajtelsmit & VanDerhei, 1997; Hawley & Fujii, 1993-1994; Palsson, 1996; Sung & Hanna, 1996a). Although there is evidence to suggest otherwise (Grable & Joo, 1997; Grable & Lytton, 1998; Wang & Hanna, 1997), it is reasonable to assume that a negative relationship exists when testing th e association between age and risk tolerance. The elder is more likely to have low level of risk tolerance; it implies that with age increasing investors have a decreasing preference for investment on risky assets. In addition to, the younger is more percentage to have high level of risk tolerance; it means that younger is less preference for investment on riskless asset than elder. (Bakshi and Chen, 1994; Brown, 1990; Dahlback, 1991; Mclnish, 1982; Morin and Suarez, 1983; Palsson, 1996; Grable, 2000; Hallahan et al., 2003; Veld and Veld-Merkoulova 2008; Frijns et al., 2008). 2.3.2 Gender A study commissioned by a major national brokerage firm found that gender is the third most powerful determinant of investing, after age and income are considered (Bajtelsmit and Bernasek, 1996). Otherwise, in some researches, gender has also been found to be an important differentiating factor in the classification of risk tolerance as females have consistently been shown to have a lower preference for risk than males (Bajtelsmit and Bernasek, 1996; Powell and Ansic, 1997; Grable, 2000). This implies that there is positive relationship between male and risky investment choice and negative relationship between female and riskless choice (Bajtelsmit and Bernasek, 1996; Blume, 1978; Hawley and Fujii, 1993 and 1994; McDermott, 1979; Rubin and Paul, 1979; Sung and Hanna, 1996; Grable, 2000; Hallahan et al., 2003; Stendardi, 2006; Veld and Veld-Merkoulova, 2008; Frijns et al., 2008). Stendardi, (2006) noted that brokers state that women investors have a distinct financial decision-making style that differs significantly from men (Worley, 1998). These brokers propose that female investors tend to be more detail oriented, they tend to want to read more about and understand their investments better, and they ask more questions. All of those authors confirm that men would have greater risk tolerance than women. Based upon those researchers and this factor, the following study in our investigate are adding into our questionnaire in socio-economic variables.
7

2.3.3 Education Level Increasing educational level attainment is associated with increased levels of risk tolerance as well, indicated by Baker and Haslem (1974), Cicchetti and Dubin (1994), Cohn et al. (1975), Masters et al. (1992), Zhong and Xiao (1995), Schooley and Worden (1996), Shaw (1996), Grable (2000), and Veld and Veld-Merkoulova (2008). It is found that individual investors with university or college education are more likely to invest in risky asset. The level of education is thought to impact on a persons ability to accept risk. Specifically, higher attained levels of education are felt to increase a persons ability to evaluate risk and are therefore considered to be positively related to higher financial risk tolerance (Baker and Haslem, 1974 Haliassos and Bertaut, 1995; Sung and Hanna, 1996). A persons level of formal education has been found to influence risk tolerance. Numerous researchers have concluded that greater levels of attained education are associated with increased risk tolerance (e.g., Baker & Haslem, 1974; Grable & Lytton, 1998; Shaw, 1996). Research findin gs tend to support the hypothesis that a positive relationship exists between educational attainment and financial risk toleran ce (Sung & Hanna, 1996a; Zhong & Xiao, 1995).One interesting issue on demographic information is whether more educated individuals have more money and so they are more likely to need the services of a financial planner and hence undertake the risk survey. This implicate that one with higher education level is more likely to take more risk tolerance than one with lower education level. 2.3.4 Income Level Over the years a positive pattern between income and financial risk tolerance has been observed. For example, Cohn, Lewellen, Lease, and Schlarbaum (1975) concluded that relative financial risk tolerance increases with wealth and income. Similar findings have been reported by Cicchetti and Dubin (1994), Friedman (1974), Income and wealth are two related factors that are hypothesized to exert a positive relationship on the preferred level of risk. Alternatively, wealthy people may be more conservative with their money while people with low levels of personal wealth may view risky investments as a form of lottery ticket and be more willing to bear the risk associated with such payoffs. Its means increasing income level of individual investor is associated with increased levels of risk tolerance (Hallahan et al., 2003; Veld and Veld-Merkoulova 2008; Grable, 1999; Grable, 2000). This empirical result is
8

consistent with earlier research (Baker and Haslem, 1974; Cicchetti and Dubin, 1994; Cohn et al., 1975; Masters et al, 1992; Schooley and Worden, 1996; Shaw, 1996; Zhong and Xiao, 1995). This implies that one with higher income level could afford more risk than one with lower income level (means the rich person will get more risk tolerance scores than poor person). For the latter and early research outcome, however, the issue is not clear cut. On the one hand, wealthy individuals can more easily afford to incur the losses resulting from a risky investment and their accumulated wealth may even be a reflection of their preferred level of risk. 2.3.5 Marital Status Marital status has also been presumed to impact on financial risk tolerance; however, the precise nature of the relationship is not clear. One points claims that unmarried person are more risk tolerant than married individuals because they have less responsibilities than married people, particularly in respect to adopted person and face less social risk than single (that is, potential loss of esteem) when undertaking risky investments (Roszkowski et al., 1993). On the other hand, it has also been suggested that married individuals have greater risk taking propensities because of a greater capacity to absorb unfavourable outcomes than unmarried person (Grable, 1999; Grable, 2000). Investigation of the investment decisions made by married individuals presents a unique challenge to researchers, as the investment portfolio of the couple may reflect the combined risk preferences of the couple (that so-called family financial invest risk), many studies have shown that marital status interacts with financial information and issues differently (Bernasek and Shwiff, 2001). Furthermore, marital status is a potentially important demographic that impacts on the preferred level of risk or risk tolerance in the investment process that single investors are more risk tolerance than married investors, although some research has failed to identify any significant relationship between risk tolerance and marital status. Increasing levels of risk tolerance also have been associated with being single are unmarried person(Baker and Haslem, 1974; Roszkowski et al., 1993; Sung et al, 2000; Hallahan et al., 2004; Veld and Veld-Merkoulova, 2008). On the other hand, it has also been suggested that married individuals have greater risk taking propensities because of a greater capacity to absorb unfavorable outcomes. The empirical research fails to provide any insights as to which of these competing theories may be valid. A number of studies have failed to
9

identify any significant relationship between marital status and financial risk tolerance in individual investors (McInish, 1982; Masters, 1989; Haliassos and Bertaut, 1995). All of these authors show the result in their research that single person could afford risk more than married person, means unmarried person have more risk tolerance scores than married person.

3. Research design In order to understand.22;;; individual investors risk tolerance, risk perceptions and socio-economic status background. The first step is to select and design our methodologies, including conjoint analysis method and survey. The second procedure is trying our best to capture information in individual investors related. The third walk, we have to sort and analysis those cross-section data that we collected. And the last pace, is according the above-mentioned, we statement those empirical results, and make conclusions about our research, involving study after this paper. 3.1 Methodology Because of our research is use two methodology as the following we showed. First methodology is survey; this method is a way we selected to catch most of information about individual investors in investment behavior and socio-economic status difference (or portfolio choice). Second methodology is conjoint analysis methodology; in order to analysis those cross-section data and find out the relation between socio-economic factors background, risk tolerance, risk perceptions and other ingredients could influence portfolio choice among individual investors. 3.1.1 Measuring Investors Risk Tolerance and Risk Perceptions The survey in this research are separate four portions, First segment is devoted to measurement of the respondents (target of individual investors) degree of risk tolerance and risk perceptions. While, risk tolerance and risk perceptions is a well-studied in examine respondents as they face or talking to invest what kind of measures, factors, and concepts will influence their choice. The next part of the questionnaire is aims to investigate the respondents investment decision on portfolio choice, according their own demographic status background, risk
10

tolerances, risk perceptions and their owe investment preference on portfolio attributes and total utility in their investments return with losses. Third section is the behavior of investors they upon are present in their portfolio choice (or financial products choice). This part are considers nine investment attributes for individual investor. In order to realize when investors faced on different investment combinations or discrepancy portfolios what kinds of attributes attract investors to make their portfolio choices? The final category is investors preference about asset management institution attributes. On this segment, the asset management choice involving seven attributes that we design to select by respondents. Under this situation, our respondents may via this or those choices in asset management institution, for this reason, those respondents could reflect their preference on asset management institution attributes. 3.1.1.1 The Designed and Structure of Questionnaire In this study, we are going to approach our target in this research; under this situation we adopted a research method. In order to cluster first-hand information abort individual investors, we used questionnaire (a survey) to collection individual investors socio-economic status differ from with previous studies (focus in fixed groups) that introduced the commercials database or panel data. The aim of this phase is to investigate the relationship between demographic factors, financial risk perceptions and financial risk tolerance in individual investors portfolio choice. Our analysis of the relationship between participant demographics background and risk tolerance reveals that gender, personal income level, family income level, marital status, education level and number of independent are significantly associated with financial risk tolerance and financial risk perceptions. We use of survey is a widely accepted method, according to Grable and Lytton (2001) for assessing financial risk tolerance score and financial risk perceptions score. Because of the complexity of the attitudinal on risk tolerance and risk perceptions construct, a sophisticated psychological testing instrument like risk measure is required (Callan and Johnson 2003). A good attitudinal test meets accepted psychological standards for both face validity (perceived relevance of the questions) and predictive validity (prediction of later performance or behavior). It also has reliability (consistency in results for repeated tests of the same person or in divide some groups). Therefore, we collection the information via the questionnaire covering four sections,
11

including (i) risk tolerance and risk perceptions of individual investor, (ii) socio-economic status of individual investor, (iii) individual investor on portfolio choice and (iv) asset management institution choice. Accompanying the risk tolerance and risk perceptions test is a set of fourteen demographic background questions; socio-economic status covers individual gender, age, personal income level, household income level, number of independent, education level and marital status. 3.1.1.2 The Item of Socio-economic Status Variables Demographic statistics data including many dimensions, socio-economic variables are among one part. Based on this foundation, we choice fourteen questions on the questionnaire in second part. We set up few decisions on six topics, gender (question 3), age (question 1), personal income level (question 6, 14), education level (question 5, 7), material status (question 11, 12, 13) and occupation (question 4). In addition to, our socio-economic status variables are involve occupations and finance management motivations (question 9). The age yeas old, more than 61 yeas old), personal income level per month and education level are to sort five stages. Financial Goals, including individual investor how to definition their investment and how to achieve their investment goals or financial goals. Sources of financial information, comprehend few channel that individual investor could obtain useful financial information what they want easily. 3.1.1.3 Measurement of Risk Tolerance and Risk Perceptions According to Grable and Lytton (2001), this test was conducted to confirm that respondents who scored lower (or higher) on one item generally scored similarly on other items. Mainly, it was determined that persons who were categorized as having low financial risk tolerances to prefer less confident in their investment behaviors or portfolio choice, less aggressive in their investing behaviors, and more likely to avoid risky financial situations on risk assets than those who were categorized into higher risk tolerance individual investors. In our risk measure, the measurements of risk tolerance level and risk perceptions score in this questionnaire. In this part, we design five questions in risk tolerance and six questions in risk perceptions, involving risk measurements, invest loss, the complexity of investment option, portfolio choice and difference possibility to get more return but those two funds in common with expected return. Besides, in this area we also investigate the risk tolerance and risk
12

perceptions of individual investors on portfolio choice. In self-assessed risk tolerance and risk perceptions for some individual investors, their answers to the survey suggested a risk tolerance different from their own perception of their ability to absorb risk, and the majority tended to underestimate their risk tolerance to varying degrees. Mainly, our results serve to confirm the rationality of individuals choices. Investors responses to individual questions, which represent their expressed preferences, are broadly consistent with their overall level of risk tolerance, that is, their revealed preference. The equation of risk tolerance and risk perceptions as following: Risk Perception / Tolerancei = 0 + 1 Agei + 2Genderi + 3 Material Statusi + 4Occupationi + 5 Income Leveli + 6 Education Leveli + i where Risk Perception / Tolerancei is involving six factors to composed risk tolerance and risk perceptions. Those factors are either socio-economics status variables. 3.1.1.4 Choice Experiment on Investment Portfolio When analyzing the investment decisions of private investors, we argue that the investment choice is essentially indifferent from any other portfolio choice. Therefore we have to design some different choice about portfolio in our portfolio choice to understand what the individual investors favorite are. In the first section, a portfolio choice is represented as a series of sequential steps: (1) definition of the choice problem, (2) generation of alternatives, (3) evaluation of attributes of alternatives, (4) choice of the optimal alternative or a favorite portfolio and (5) implementation of the choice. Specifically, we designate the different weight to construct the portfolio choice combined with three categories of investment instruments as follows: stock (as high-risk asset), bond (as low-risk asset) and saving account (as risk-free asset). All of those investment profiles are used in the different portfolio choice experiment that individual choired that we present in table 4. Following Hallahan et al. (2003), risk tolerance measure is defined as the amount of uncertainty or the volatility of investment return that an investor is willing to accept when making a financial decision. Therefore, this equation is the influence factors of portfolio choice effected: (1)

13

Ch oice U Portfolio = 0 +1 Attribu tes Portfolio + 2 R isk Perception s

+3 R isk T oleran ce + 4S E S +5 R isk M easu res + 6 AM + 7 In form ation +8Goals +i

(3)

Ch oice where U Portfolio is display the interaction relationship between investors demographic

backgrounds and other four independent variables. R isk T oleran ce is stands for how many risk (or loss) that individual investors could and willing to bear and its combined the investors demographic background and their dependent factors. R isk Perceptions are present for risk perceptions that are about risk and reflecting a persons private attitude towards taking on risk, is a complex psychological concept.

Asset Managem ent means asset management properties, here are seven attributes can
afford investors selection, this method is trying to understand what the key element is influenced choice when investors on asset management institution; i is represents for the error term.

3.1.2 Random Utility


In order to understand how individual investor choose their preference portfolio, we need to determinate which utilities are they most preference. Maximum utility is achieved by many factors that individual investor they preference in various financial products of utilities. An interaction term is used because the utility of investments such as portfolio attributes and other variables, for instance, socio-economics status, risk tolerance and risk perceptions. Most of those dependent variables are existence random utilities. If those random utilities could make interaction between themselves, therefore the totally utility of individual investor may obtain. Individual investor may accord their utility of investment item or investment preference to choose portfolio or make an investment plan. Depend upon on this reason; individual investor may irrationality on their decision of portfolio choice.

3.1.3 Conjoin Analysis


Conjoint analysis, introduced by Green and Rao (1971), has developed into a primary marketing technique that measures tradeoffs between multi-attribute products and services
14

(Bradlowand Marshall, 2002). As implied by the previously discussed choice framework and limitation, we need to define a utility functions that expresses investors preferences for different investment opportunities and incorporates various attributes and investors characteristics. We use a conjoint analysis choice modeling concept to model this utility. A conjoint analysis is utilized to model the utility for portfolio choice expressing investors preferences for different portfolio choices and mingle various portfolio attributes and individual investors personalities. Conjoint analysis examines how individuals derive preferences for products or services by assuming that they sum individual utilities we called total utility, for each attribute to obtain the overall utility for a product or service. Total utility of an investment is a function that combines the number of utilities each alternative has. The investor chooses the investment combination that offers highest utility. Given the different preferences, we can find the probability of selecting one product over another one. As implied by the previously discussed choice framework, we need to define a utility function that expresses investors preferences for different investment opportunities and incorporates various attributes and investors characteristics. We use a conjoint choice modeling concept to model this utility. The investor evaluates which investment opportunity has the highest utility based on a series of attributes that reflect the investors investment approach, and evaluates each attribute and assigns a certain amount of utilities (measurement unit of utility) to that attribute. This approach could examine how individual investors derive their preferences for financial assets by assuming that they sum individual utilities for each attribute to obtain the overall utility for financial assets. Given the different preferences, we can find the probability of selecting one financial asset over another one. The conjoint framework is aimed to estimate the probability that an investor set a certain amount in a particular investment portfolio. The choice sets are based on a simplified financial market environment.

3.2 Survey Procedures


A decisive part of any conjoint analysis research is the design and piloting of the risk tolerance questionnaire or examine risk measure instrument (Poynter, 1999, Veld and Veld-Merkoulova, 2008). The preparatory qualitative study (discussed earlier) and previous study in investment decision making identified appropriate portfolio attributes and risk levels. The questionnaire was tested on a small sample of business student to check for difficulties in completion or unusual results prior to undertaking and understand the main study but no
15

problems were found from those tested sample. Respondents were pre-qualified by the completion and return of a form indicating their willingness to participate our investigation or ready to take part in financial market. Those who indicated willingness were sent a questionnaire on paper in this research. In a small number of cases responses were collected in a one-to-one interview rather than. It was recognized that dramatic changes in financial investment market conditions could have an impact on the responses received and their choice. However, the socio-economic background information was collected from January to February 2009, a period during which there was no dramatic change in financial investment market conditions.

3.2.1 Data Collection


Our data collection would be done in the form of a survey that exposed respondents to different choice sets with a number of alternatives each. Comparing with previous studies, most of researches are limited their sample collection and choice, but in our article no restrictions are imposed on the target population in terms of educational levels, gender, income level, material status, age, number of independent, investment goals, sources of investment information, family income source and occupation. As such the cross-section of respondents (individual investors) best reflects the difference in individual investors. The sample units have been choice by means of a quasi-randomized sample procedure.

3.2.2 Data Analysis


Descriptive statistics were obtained for the socio-economic background data and responses to the risk tolerance and risk perceptions survey. For the socio-economic background information, percentage for each individual investor on risk tolerance and risk perceptions responses interacted were computed. For the background data from the risk tolerance and risk perceptions on portfolio choice survey. Four risk measures are involving, semi-variance, expected value of loss, probability of loss and standard deviations were used to examine each attributes level to gain a feel for individual investor respondents. The relative importance of each attributes of each attributes level were calculated to examine individual investor preferences between attributes on portfolio choice.

4. Empirically Results
16

Analysis of the dataset has five facets: investigating the relationship among socio-economic factors, risk tolerance, risk perceptions, asset management institution choice and portfolio choice. Heterogeneity of individual investor in risk tolerance and socio-economic status largely determines portfolio choice. The empirically result are display in appendix from table six to thirteen. The potential information is our questionnaire is involving two sections, both adopted scores. For some reasons that we desire to understand what risk measure or measures will influenced investors invest behavior. In our study on socio-economic status (or factors) are contain ten variables, gender, age, occupation, education level, personal income level, marital status, number of independent persons, household income source, investment goals and source of investment information. According to our data, we can analysis the relationship is positive or negative on significant condition. All of our researches empirical results are display in the following tables from table 5 to table 19.

4.1 Descriptive Statistics 4.1.1 Socio-Economics Background


Although in different data collection methods but the target population consists of individuals who have invested in financial market or could and planning to invest in financial markets are no limitations in our samples. The surveys in our cross-sectional data are sending out 410 questionnaires, the questionnaires recycle percentage on my investigating is 88%. Under this situation, the percentage of effective questionnaire is almost 79%. Therefore, we can calculate the useless and unrecyclable questionnaires are 125; the ratio is near to 30.49%. Using a cross-sectional data from four parts of 285 questionnaires that we obtained by questionnaire are composed 3420 observations. The purpose of our design is to represent individual investor has different choice in questionnaire, one portfolio contains three investment combinations, it is means one questionnaire of portfolio are involving twelve investment combination (or investment opportunities) but stand for various choice condition. The information of effective samples (questionnaires) are on distribution in descriptive statistics are present in the table 5. First factor is gender, the observations on male and female are 1,718 and 1,692 and the ratio is 50.53% and 49.47%. In order to simplify the sentence, we will change the presentation
17

method into observations (percentage). Second element is age, to discriminate the age on four ranges, observations on age are 1572 (45.96%)852 (24.91%)576 (16.84%) and 420 (12.28%). The third socio-economic that we differentiate between in eight pools are profession, observations are 516 (15.09%), 768 (22.46%), 156 (4.56%), 624 (18.25%), 144 (4.21%), 900 (26.32%), 84 (2.46%), and 228 (6.67%). Fourth is aim to education level, we are going to make off 4 ranks, observations are 108 (3.16%), 516 (15.09%), 1,872 (54.74%) and 924 (27.02%). Fifth in socio-economic is focus on personal income level per month, identify income five grades and the observations are 1080 (31.58%)480 (14.04%)744 (21.75%) 684 (20%) and 432 (12.63%). Sixth, we differentiate marital status to three states, marriedunmarried and others, observations are 2052 (60%)1332 (38.95%) and 36 (1.05%). In seventh is number of independent in this survey, we separate no dependent and lest one person, observations are 2460 (71.93%) and 960 (28.07%). Eighth, we discriminate family income source three statuses, independent; single salary and double salaries, the observations of family income source 2076 (60.7%), 156 (4.56%) and 1188 (34.74%). The ninth socio-economic factor is the set up that we design six investment goal of individual investors, observations are 936 (27.37%)1272 (37.19%)756 (22.11%)348 (10.18%)60 (1.75%) and 48 (1.4%). The last socio-economic issue is the source of investment information have five sources, observations are 914 (27.03%)1392 (40.71%)731 (21.38%)252 (7.37%) and 120 (3.51%).

4.1.2 Risk Attitude of Individual Investor on Portfolio Choice


The design of our survey for individual investors to examine their financial risk attitudes when they are confront uncertain financial risk and trying to investment. Especially, when individual investor has invested in financial marketplace, they may bear loss or catch profit on their investment plan or process, the results of examination are present in table 16. In addition to, our research results are fit with our expectation before we mentioned in literature review. For instance, the reverse of risk tolerance and reverse of risk perceptions are more preference on return of bond investment than moderate of risk tolerance and seeking of risk tolerance is stand for in this portfolio attributes, investors are dislike risk and deflection conservative. According to the above-mentioned, we are going to calculate the investment damage that how much money loss they are willing to accept or they can afford through the portfolio
18

we design. After the data we obtained, among three risk attitudes of individual investor they are no preference investment in cross country. Besides, in those groups that we divide from our all samples, have special partiality in the weight of save accounting, means those investors are deviation moderate investment attitudes more than other individual investors. But in seeking of risk tolerance and seeking of risk perceptions present negative relationship with weight of save accounting, representative investors in this group has less preference in safety asset, they are not partials to reverse or moderate than other individual investors.

4.2 The Impact of Socio-Economic Status and Risk Tolerance on Investors Portfolio Choice
This part is focus on the impact of risk tolerance and socio-economic status on portfolio choices. First, the portfolio attributes in return of bond investment is display negative, its mean investors less preference in return of bond investment; weight of save accounting is appear significant show positive indicate investors more preference in weight of save accounting this financial product. Second, the socio-economic status factors in gender and adopted person are appearing negative and significant; its signifying investors are less risk tolerance on portfolio choices. Third part is financial related background; financial professional training and certificate of financial management are appearing negative relationship on portfolio choice. In addition to, major in finance are show positive relationship on portfolio choice. In fourth and fifth sections, in trading of investment goal and semi-variance of risk measures are both remarkable. In trading of investment goal is positive but in semi-variance of risk measures negative, stand for those two factors present different directions on portfolio choice. The most important is the relationship between risk tolerance and socio-economic status. Most of risk tolerance represents outstanding but two risk tolerance issues is positive and the other one is exhibit negative, stand for individual investors are focus on financial and investment risk. Overall, these results suggest that gender, tertiary education, income, and wealth are all related to risk tolerance.

4.2.1 Gender
First demographic is gender, the empirically results are presenting in table 9. In our estimation data that the noticeable factors are fifteen factors. Those three attributes as the
19

following: total variance of portfolio; total standard deviation of portfolio and total expected loss of portfolio in male and female are noticeable. Means those three variables will make influence for investors when they are in portfolio choices. Portfolio choice is outstanding in male, so that we can realize those two variables have positive effect on portfolio choice in individual investors. Previous author is frequently argued to determine risk tolerance is gender, and Bajtelsmit & Bernasek (1996), Palsson (1996), Jianakoplos and Bernasek (1998), Bajtelsmit, Bernasek and Jianakoplos (1999), Powell and Ansic (1997), and Grable (2000) find support for the notion that females have a lower preference for risk than males. Summary in gender, female investor would show lower risk tolerance than male one for risky assets.

4.2.2 Personal Income Level


Second demographic is income level, the empirically in our data collection between income level and risk tolerance, while this relationship is significantly different for both high income level and low income level on individual investors. The empirically is presenting in table 10. Income level and risk tolerance are two related factors in demographic background in other investment factors that are exist positive or negative relationship on the preferred level of risk tolerance (see Friedman, 1974; Cohn, Lewellen, Lease & Schlarbaum, 1975; Blume, 1978; Riley & Chow, 1992; Grable & Lytton, 1999; Schooley & Worden, 1996; Shaw, 1996; Bernheim et al., 2001). High income levels are more significant factors than low income level on portfolio choices. For example, the gender; number of dependent; weight of save accounting; weight of bond investment; return of risk-free saving; return of bond investment on high income level are existing negative relationship, indicated that will decrease investment desire on portfolio choice. But the total expected value of portfolio is positive; imply in this factor the investor willing to investment on portfolio.

4.2.3 Education Level


One interesting issue that our data allows us to explore is the relationship between education and risk tolerance. The empirically is presenting in table 11. For this generation, a university degree was far less common compared to current trends and at the time, the general perceptions was that a tertiary education would secure a financial future. Our data enables us to explore the relationship between respondents answers to their education level of risk
20

tolerance and completed education. For example, low education level of weight of stock investment; total variance of portfolio; income; gender; financial professional training; negative relationship, but in age variable is positive relationship with risk tolerance. Another education level is high, and return of bond investment; certificate of financial management are negative relationship and the weight of save accounting is positive relationship with risk tolerance. Education level is a third demographic factor that is thought to increase a persons capacity to evaluate risks inherent to the investment process and therefore endow them with a higher financial risk tolerance (see Baker & Haslem, 1974; Haliassos & Bertaut, 1995; Sung & Hanna, 1996).

4.2.4 Marital Status


This socio-economic variable may have some research or conclusions in previous authors studies. This factor has negative relationship between For instance, Grable and Lytton (1998); Grable (2000); Grable and Lytton (1999), their conclusion are present positive relationship with risk tolerance on portfolio choices. This is why individual investor are more risk tolerance when they on investing, because a lot of life pressures let those investors have to bear risk. They cant loss a lot of money when they are invest, so under the situation that they afford loss they have to wait for a long time to avoid capital loss, this is why they are have positive relationship about risk tolerance and portfolio choice. In another side, Hallahan et al. (2003) is to address negative relationship with risk tolerance on portfolio choices. Therefore, this empirically result is trying to illustrate the lower risk tolerance scores on portfolio choice in our research, it is indicated investor may have less interesting on portfolio choice.

4.3 The Impact of Socio-Economic Status and Risk Perception on Investors Portfolio Choice
This part is aim to the impact of risk perceptions and socio-economic status on portfolio choices. First, the portfolio attributes in return of bond investment is display negative, its mean investors less preference in return of bond investment; weight of save accounting is appear significant show positive indicate investors more preference in weight of save accounting this financial product. Second, the socio-economic status factors in gender and adopted person are appearing negative and significant; its signifying investors are less risk tolerance on portfolio choices.
21

Third part is financial related background; financial professional training and certificate of financial management are appearing negative relationship on portfolio choice. In addition to, major in finance are show positive relationship on portfolio choice. In fourth and fifth sections, in trading of investment goal and semi-variance of risk measures are both remarkable. In trading of investment goal is positive but in semi-variance of risk measures negative, stand for those two factors present different directions on portfolio choice. The most important is the relationship between risk perceptions and socio-economic status. Most of risk perceptions represents outstanding but two risk perceptions issues is positive and the other one is exhibit negative, stand for individual investors are focus on financial and investment risk. Overall, these results suggest that gender, tertiary education, income, and wealth are all related to risk perceptions. The empirically result are show in table 8.

4.3.1 Gender
First, the portfolio attributes in international investment; total expected loss of portfolio; choice of asset management choice consulting and major in finance; standard deviation are display negative relationship, its mean investors less preference in international investment; total expected loss of portfolio; choice of asset management choice consulting and major in finance is appear significant appear negative relationship indicate investors less preference in those variables on portfolio choice. Another side in total standard deviation of portfolio; adopted person; the relatives are positive relationship with risk perceptions, means investors more investment preference on portfolio choice. The most importance relationship in this section is risk perceptions and gender. In male side, two risk perceptions are positive and two risk perceptions are negative. But the female is different that two risk perceptions are positive and only one risk perceptions is negative, its means female may understand more risk perceptions on our samples. The empirically is presenting in table 10.

4.3.2 Personal Income Level


Second demographic is income level, the empirically in our data collection between income level and risk perceptions while this relationship is significantly different for both high income level and low income level on individual investors. The empirically is presenting
22

in table 12. Income level and risk perceptions are two related factors in demographic background (first is adopted person in male; second is gender in female, they are opposite) in other investment factors that are exist positive or negative relationship on the preferred level of risk perceptions. High income levels are more significant factors than low income level on portfolio choices. For example, the gender; adopted person; weight of save accounting; weight of bond investment; return of risk-free saving; return of bond investment; the expected value of portfolio on high income level are existing negative relationship, indicated that will decrease investment desire on portfolio choice. But the investment goals of trading are positive; imply in this factor the investor willing to investment on portfolio.

4.3.3 Education Level


According to our research and analysis, some interesting finding in this issue that our data allows us to explore is the relationship between education and risk perceptions. The empirically is presenting in table 14. For this generation, a university degree was far less common compared to current trends and at the time, the general perceptions was that a tertiary education would secure a financial future. Our data enables us to explore the relationship between respondents answers to their education level of risk perceptions and completed education. For example, low education level of weight of stock investment; total variance of portfolio; income; gender; financial professional training; are exist negative relationship, but in age variable is positive relationship with risk tolerance. Another education level in high and weight of saving accounting is positive but certificate of financial management are negative relationship with risk perceptions. Education level is a third demographic factor that is thought to increase a persons capacity to evaluate risks inherent to the investment procedure and therefore endow them with higher financial risk perceptions in individual investors on portfolio choices or financial investment decision.

4.3.4 Marital Status


After we investigate about marital status of investors and interaction with portfolio choice and risk perceptions, some findings are interesting. According to earlier statement and comparing with the relationship between risk tolerance and marital status. In addition to, we can discover among risk tolerance and risk perceptions are present negative relationship in
23

marital status. The result that we show in the lower risk perceptions on portfolio choice means marital status will decrease investment desire the risk perceptions on portfolio choice.

5. Conclusions Identifying key factors influencing individual investors decision to make portfolio choices is of importance to understand their heterogeneous investment behavior, however previous research based on experimental evidence is extremely space especially in terms of perspective from individuals perceptions on risk tolerance and risk perception. This paper explores individual investors preference for portfolio choices and empirically investigates impacts of risk tolerance and risk perception on their investment decision. Specifically we decompose socio-economic stats difference in investment preference for portfolio choices with respect to investors gender, age, income level, marital status, education level and household income level. Using conjoint analysis on investment experiments to obtain some evidences from a sample of 285 respondents in survey, our results indicate that investors decisions to make their portfolio choices are significantly and negatively related to personal labor income level, female and married investors, and square of overall scores on risk perception while however the case is positively associated with their household income level and the level of risk tolerance. This finding implicates that investor with higher risk tolerance level shows higher likelihood to make their investment decision on portfolio choices. Moreover, individual investors would like to make the investment decision on portfolio choices with higher weights of bond investment and saving. Interestingly, they do not prefer to make portfolio choices with international portfolios and short-run investment horizon. Besides, it is found that male investor demonstrates much preference on portfolio choices with higher percentage of total return.

24

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Annaert, J., Ceuster, D., Marc, J.K., and Hyfte, W.V., 2005. The value of asset allocation advice: Evidence from the Economists quarterly portfolio poll. Journal of Banking and Finance 29, 661680. Bakshi, G.S., and Chen, Z., 1994. Baby boom population aging, and capital markets. Journal of Business 67(2), 165-202. Basu, S., Raj, M., and Tchalian, H., 2008. A Comprehensive Study of Behavioral Finance. Journal of Financial Service Professional July, 51-63. Bajtelsmit, V., L., Bernasek, A., and Jianakoplos, N., A., 1999. Gender differences in defined contribution pension schemes. Financial Services Review, 8, 110. Clark-Murphy, M., and Soutar, G.N., 2004. What individual investors value: Some Australian evidence. Journal of Economic Psychology 25, 539555. Coleman, S., 2003. Woman and risk: An analysis of attitudes and investment behavior. Academy of Accounting and Financial Studies Journal, Volume 7, Number 2, 99-115 Diacon, S., and Hasseldine J., 2007. Framing effects and risk perception: The effect of prior performance presentation format on investment fund choice. Journal of Economic Psychology 28 3152. Farley, D., E., 2000. Achieve a balance between risk and return. Healthcare Financial Management June, 54-58. Frijns, B., Koellen, E., and Lehnert, T., 2008. On the determinants of portfolio choice. Journal of Economic Behavior and Organization 66, 373386. Faff, R., Mulino, D., and Chai, D., 2008. On the linkage between financial risk tolerance and risk aversion. The Journal of Financial Research 31(1), 123. Grable, J., and Lytton, R.H., 1999. Financial risk tolerance revisited: the development of a risk assessment instrument. Financial Services Review 8, 163181. Grable, J., and Lytton, R.H., 1999. Assessing financial risk tolerance: Do demographic, socioeconomic and attitudinal factors work? Journal of FRHD/FERM Divisions of A AFCS, 1-9 Grable, J. E., 2000. Financial risk tolerance and additional factors that affect risk taking in everyday money matters. Journal of Business and Psychology 14(4), 625-630. Grable, J., and Lytton, R.H., 1998. Investor risk tolerance: Testing the efficacy of demographics as differentiating and classifying factors. Financial Counseling and Planning 9(1), 61-70. Grable, J. E., and Joo, S., 2000. A Cross-Disciplinary examination of Financial Risk Tolerance. Consumer Interests Annual 46, 151-157 Grable, J., and Lytton, R.H., 2001. Assessing the concurrent validity of the scf risk tolerance question. Financial Counseling and Planning, volume 12(2), 43-53. Hallahan, T., Faff, R., and McKenzie, M., 2003. An exploratory investigation of the relation between risk tolerance scores and demographic characteristics. Journal of Multinational Financial Management 13, 483-502. Hallahan, T., Faff, R., and McKenzie, M., 2004. An empirical investigation of personal financial risk tolerance. Financial Services Review 13, 5778. Holt, C., A., and. Laury, S., K., 2002. Risk Aversion and Incentive Effects. The American economics review December, volume 92, No.5, 1644-1655. Keller, C., and Siegrist, M., 2006. Investing in stocks: The influence of financial risk attitude and values-related money and stock market attitudes. Journal of Economic Psychology, 27, 285303. Kim, Kenneth A., and Nofsinger, John R., 2008. Behavioral finance in Asia. Pacific-Basin Finance Journal 16 17.
25

12.
13. 14. 15. 16. 17. 18. 19. 20. 21.

22. Morin, R.A., and Suarez, A.F., 1983. Risk Aversion Revisited. The Journal of Finance 38(4), 1201-1216. 23. MacGregor, D.G., Slovic, P., Berry, M., and Evensky, H., R., 1999. Perception of Financial Risk: A survey study of advisors and planners. Journal of Financial Planning 12(8), 68-86. 24. Nagy, R.A., and Obenberger, R.W., 1994. Factors influencing individual investor behavior. Financial Analysts Journal 50(4), 63-68. 25. Olson, D., and Bley J., 2008. Asset allocation with differential borrowing and lending rates. International Review of Economics and Finance, 17, 629643. 26. Riley Jr., W., B., and Victor C., K., 1992. Asset allocation and individual risk aversion. Financial Analysts Journal 48(6), 32-37. 27. Ricciardi, V., and Simon, Helen K. 2000. What is behavioral finance? Business, Education and Technology Journal Fall, 1-9. 28. Ricciardi, V., 2006. A research starting point for the new scholar: a unique perspective of behavioral finance. ICFAI Journal of Behavioral Finance, 3(3), 6-23. 29. Roszkowski, M. J., Snelbecker, G. E., & Leimberg, S. R. (1993). Risk tolerance and risk aversion. In: S. R. Leimberg, M. J. Satinsky, R. T. Leclair, & R. J. Doyle (Eds.), The Tools and Techniques of Financial Planning (4th ed.) (pp. 213~225). Cincinnati: National Underwriter. 30. Stendardi, E., J., Graham J.F., and OReilly, M., 2006. The impact of gender on the personal financial planning process: Should financial advisors tailor their process to the gender of the client? Humanomics Vol. 22 No. 4, 223-238. 31. Schooley, D., K. and Worden D., D., 2001. Risk aversion measures: Comparing attitudes and asset allocation. Financial Services Review, 5(2): 87-99. 32. Subrahmanyam, A., 2007. Behavioural finance: A review and synthesis European Financial Management, Vol. 14, No. 1, 1229. 33. Veld, C., Veld-Merkoulova, Y., 2008. The risk perceptions of individual investors. Journal of Economic Psychology 29 226252. 34. Wang, X.L., Shi, K., and Fan, H.X., 2006. Psychological mechanisms of investors in Chinese Stock Markets. Journal of Economic Psychology 27, 762-780. 35. Yao, R., and Human S., D, 2005. The effect of gender and marital status on financial risk tolerance. Journal of Personal Finance Volume 4, Issue 1, 66-85.

Table 1 Empirical results from other studies


NoteThe letter after year means the dependent variable. (PC=PORTFOLIO CHOICE and RT=RISK TOLERENCE, RISK MEASURES=RM). Explanatory variable Positive relation with Negative relation with No relation with Socio-Economic Status Age Grable (2000)RT Frijns et al. (2008) PC Morin and Suarez (1983) RT Riley Jr. and Chow (1992)PC Grable and Lytton (1998)RT Grable and Lytton (1999)RT Hallahan et al. (2003)RT Veld and Veld-Merkoulova (2008) RM

26

Race-Whites Race-Non-Whites Gender-Male

Grable and Lytton (1998)RT Grable and Lytton (1998) RT Riley Jr. and Chow (1992)PC Grable and Lytton (1998)RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003RT Yao and Human (2005RT Stendardi et al. (2006)PC Frijns et al. (2008) PC Veld and Veld-Merkoulova (2008) RM

Riley Jr. and Chow (1992) Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT

Gender-Female

High-Income

Morin and Suarez (1983) RT Riley Jr. and Chow (1992)PC Grable and Lytton (1998) RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT Veld and Veld-Merkoulova (2008) RM

Riley Jr. and Chow (1992)PC Grable and Lytton (1998)RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT Yao and Human (2005) RT Stendardi et al. (2006)PC Frijns et al. (2008) PC

Low- Income

Riley Jr. and Chow (1992)PC Grable and Lytton (1998)RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT

27

Table 1 (Continued)
NoteThe letter after year means the dependent variable. (PC=PORTFOLIO CHOICE and RT=RISK TOLERENCE) Explanatory variable Socio-Economic Status Marital Status-Single Positive relation with Hallahan et al. (2003)RT Negative relation with Grable and Lytton (1998) RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT No relation with Riley Jr. and Chow (1992)

Marital Status-Married High-Education

Grable and Lytton (1998) RT Grable (2000)RT Grable and Lytton (1999)RT Riley Jr. and Chow (1992)PC Grable and Lytton (1998) RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT Veld and Veld-Merkoulova (2008) RM

Low-Education

Riley Jr. and Chow (1992)PC Grable and Lytton (1998)RT Grable and Lytton (1999)RT Grable (2000)RT Hallahan et al. (2003)RT Hallahan et al. (2003)RT Hallahan et al. (2003)RT Grable and Lytton (1998) RT Grable (2000) PC Grable and Lytton (1998) RT Hallahan et al. (2003)RT

Occupation Number of Independent High-Financial Knowledge Low-Financial Knowledge Risk Tolerance

Grable and Lytton (1998) RT Grable (2000)RT Wang et al. (2006)PC Grable and Lytton (1999)RT Grable (2000)RT Grable and Lytton (1999)RT Grable and Joo (2000)RT Farley (2000)RT Grable and Lytton (2001)RT Holt and. Laury (2002) PC Hallahan et al. (2004) Faff et al. (2008)PC Frijns et al. (2008) PC Riley Jr. and Chow (1992)PC Bajtelsmit et al. (1999) PC MacGregor et al. (1999) PC Schooley and Worden (2001)PC Clark-Murphy and Soutar (2004)RM Keller and Siegrist (2006) PC Olson and Bley (2008) PC Veld and Veld-Merkoulova (2008) RM

Grable (2000)RT Grable and Lytton (1999)RT

Risk Perceptions

28

Table 2 Variables Definition and Description


Variables
Portfolio Attributes International Portfolio Investment Time Horizon Return on Bond Investment Stock Investment Risk-Free Saving Weight of Stock Investment Bond Investment Risk-Free Saving Total Return on Portfolio Measures in Percent (%)

Description
We construct the portfolio involving international investment. How long are individual investors willing to invest their portfolio? There are following two category of investment time horizon: less than 1 years and less than 7 years Return on bonds investment Return on stocks investment. Return on fixed income form saving account Investment weights of bonds investment over portfolio Investment weights of stocks investment over portfolio Investment weights of saving account over portfolio

Total return of portfolio in percent is calculated by summed with return on bonds, stocks and saving account, respectively. Measures in Amount (NT$) Total return of portfolio in amount is calculated by summed with return on bonds, stocks and saving account, respectively. Total Expected Value of Portfolio This is calculated by summed with expected return on bonds, stocks and saving account, respectively. Total Variance of Portfolio This is generated by summed with variance of return on bonds, stocks and saving account, respectively. Total Standard Deviation of Portfolio This is calculated by summed with standard deviation of return on bonds, stocks and saving account, respectively. Total Expected Loss of Portfolio This is estimated by summed with expected loss from bonds, stocks and saving account, respectively. Private Wealth Management Service If individual investor follows private wealth management adviser to investment, we set up the dummy variable to be 1 and 0 otherwise. Socio-Economic Status Gender 1=male investor ; 2= female investor Age (Years Old) 1=21~30 years old ; 2=31~40 years old ; 3=41~50 years old; 4=51~70 years old Occupation 1=finance ; 2=goverment service ; 3=self-emplyment; 4=service; 5=manufacturing; 6=student; 7=unemployment ; 8=others Education (Level) 1= less than junior high school; 2= senior high school; 3=university; 4= Master or Ph.D. Number of Children and the Elder The number of children (less than 18 years old) and the elder (more than 65 years old) in a houshold. Material Status 1=married; 2=devoice; 3=single; 4=other Personal Labor Income Level per 1=NT$10,000~NT$20,000; 2=NT$20,001~NT$30,000; 3=NT$3,001~NT$40,000; Month (NT$) 4=NT$40,001~NT$50,000; 5=more than NT$50,000 Household Income Level per Month 1=less than NT$20,000; 2=NT$20,001~NT$40,000; 3=NT$40,001~NT$60,000; (NT$) 4=NT$60,001~NT$80,000; 5=NT$80,000~NT$100,000; 6=more than NT$100,001 Investmenst Characteristics Investment Goals The goal for individual investment portfolio is the following: 1=hedging; 2=increase wealth; 3=trading; 4=earning higher return; 5=diversifying Risk; 6=preferring fixed income Information Sources of Investment The main channels for individual investor to gain the investment information are as follows: 1=the relatives or friends; 2=television or radio; 3=internet; 4=financial advisors; 5=news paper or magazines.

29

Table 3 Investment Attributes and Attribute Levels Used for Choice Experiment
Investment Attributes International Portfolio Investment Time Horizons Return of Bond Investment Return of Stock Investment Return of Risk-Free Saving Weight of Stock Investment Weight of Bond Investment Weight of Risk-Free Saving Total Return of Portfolio in Percent (%) Attribute levels Oversea Investment and Domestic Investment 2% and 3.75% 14% and 20% 2.5% and 3% 0%, 50%, 100% 0%, 50%, 100% 0%, 50%, 100% 1. 2% 2. 2.5% 3. 3% 4. 3.13% 5. 3.75% 6. 8.88% 7. 11.1% 8. 14% 9. 20% 1. NT$ 200 2. NT$ 250 3. NT$ 300 4. NT$ 313 5. NT$ 375 6. NT$ 888 7. NT$ 1,110 8. NT$ 1,400 9. NT$ 2,000 1. NT$ 10,200 2. NT$ 10,250 3. NT$ 10,300 4. NT$ 10,313 5. NT$ 10,375 6. NT$ 10,888 7. NT$ 11,110 8. NT$ 11,400 9. NT$ 12,000 1. 0.0013% 2. 0.0128% 3. 0.014% 4. 0.0146% 5. 0.02% 6. 0.027% 7. 0.56% 1. 0.118322% 2. 0.036056% 3. 3.6% 4. 1.21% 5. 0.4719% 6. 1.632% 7. 0.07483% 1. NT$ 0 2. NT$-50 3. NT$-62 4. NT$-100 5. NT$-200 6. NT$-700 7. NT$-1,050 8. NT$-1,400 9. NT$-2,000 Explanation and Calculation Oversea: The investment behaves belong Taiwan financial market. Domestic: The investment behaves belong Taiwan financial market. Less 1 years and Less 5 years Oversea=3.75%; Domestic= 2% Oversea: 14%; Domestic: 20% Oversea: 2.5%; Domestic: 3%

Total Return of Portfolio in Amount (NT$)

Total Expected Value of Portfolio

Total Variance of Portfolio

1. 2. 3. 4. 5. 6. 7. 8. 9. 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. 2. 3. 4. 5. 6. 7. 1. 2. 3. 4. 5. 6. 7. 1. 2. 3. 4. 5. 6. 7. 8. 9.

[(NT$ 10,200-NT$ 10,000)/NT$ 10,000]*100%=2% [(NT$ 10,250-NT$ 10,000)/NT$ 10,000]*100%=2.5% [(NT$ 10,300-NT$ 10,000)/NT$ 10,000]*100%= 3% [(NT$ 10,313-NT$ 10,000)/NT$ 10,000]*100%=3.13% [(NT$ 10,375-NT$ 10,000)/NT$ 10,000]*100%=3.75% [(NT$ 10,888-NT$ 10,000)/NT$ 10,000]*100%=8.88%, [(NT$ 11,110-NT$ 10,000)/NT$ 10,000]*100%=11.1% [(NT$ 11,400-NT$ 10,000)/NT$ 10,000]*100%=14% [(NT$ 12,000- NT$ 10,000)/NT$ 10,000]*100%=20% NT$ 10,200-NT$ 10,000= NT$ 200 NT$ 10,250-NT$ 10,000= NT$ 250 NT$ 10,300-NT$ 10,000= NT$ 300 NT$ 10,313-NT$ 10,000= NT$ 313 NT$ 10,375-NT$ 10,000= NT$ 375 NT$ 10,888-NT$ 10,000= NT$ 888 NT$ 11,110-NT$ 10,000= NT$ 1,110 NT$ 11,400-NT$ 10,000= NT$ 1,140 NT$ 12,000-NT$ 10,000= NT$ 2,000 NT$ 10,000*(1+2%)= NT$ 10,200 NT$ 10,000*(1+2.5%)= NT$ 10,250 NT$ 10,000*(1+3%)= NT$ 10,300 NT$ 5,000*(1+2.5%)+ NT$ 5,000*(1+3.75%)= NT$ 10,313 NT$ 10,000*(1+3.75%)= NT$ 10,375 NT$ 5,000*(1+3.75%)+ NT$ 5,000*(1+14%)= NT$ 10,888 NT$ 5,000*(1+2%)+ NT$ 5,000*(1+20%)= NT$ 11,110 NT$ 10,000*(1+14%)= NT$ 11,400 NT$ 10,000*(1+20%)= NT$ 12,000 2%*(11.1%-10%)2+20%*(11.1%-10%)2=0.0013% 2%*(2%-10%)2=0.0128% 2*(2.5%-10%)2+3%*(2.5%-10%)2=0.014% 3.75%*(3.75%-10%)2=0.0146% 3.75%*(8.88%-10%)2+14%*(8.88%-10%)2=0.02% 3.75%*(3.12%-10%)2+2.5%*(3.12%-10%)2=0.027% 14%*(14%-10%)2=0.56%
0.014% =0.118322% 0.0013% =0.036056% 2 1.28% =3.6% 2 0.0146% =1.21% 2 0.00222% =0.4719% 2 0.027% =1.632% 2 0.0056% =0.07483% NT$10,250-NT$10,250=NT$0 NT$10,250-NT$10,300=NT$-50 NT$ 10,313-NT$10,375=NT$-62 NT$10,200-NT$10,300=NT$-100 NT$9,800-NT$10,000=NT$-200 NT$4,300-NT$5,000=NT$-700 (NT$4,000-NT$5,000)+(NT$5,000-NT$5,050)=NT$-1050 NT$8,600-NT$10,000=NT$-1,400 NT$8,000-NT$10,000=NT$-2,000
2
2

Total Standard Deviation of Portfolio

Total Expected Loss of Portfolio

30

Table 4 Investment Profiles Used in the Portfolio Choice Experiment


Portfolio Attributes International Portfolio Investment Time Horizons Return on Investment Instruments Weight of Investment Instruments Percentage of Return on Portfolio Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Portfolio Attributes International Portfolio Investment Time Horizons Return on Investment Instruments Weight of Investment Instruments Percentage of Return on Portfolio Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Portfolio Attributes International Portfolio Investment Time Horizons Return on Investment Instruments Weight of Investment Instruments Percentage of Return on Portfolio Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Portfolio Attributes International Portfolio Investment Time Horizons Return on Investment Instruments Weight of Investment Instruments Percentage of Return on Portfolio Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Profile 1 Domestic Less 5 year Bond=2% Bond=100% 2% (NT$200) NT$10,200 0.0128% 1.13% NT$-100 Profile 5 Oversea Less 1 year Bond=14% Bond=100% 14% (NT$1,400) NT$11,400 0.56% 7.48% NT$ -1,400 Profile 9 Oversea Less 1 year Bond=3.75% Stock=14% Bond=50% Stock=50% 8.88% (NT$888) NT$10,888 0.00222% 0.4719% NT$ -700 Profile 13 Oversea Less 5 year Stock=2% Stock=100% 2% (NT$200) NT$10,200 0.0128% 1.13% NT$ -200 Profile 2 Domestic Less 1 year Risk-Free Saving=20% Risk-Free Saving=100% 20% (NT$2,000) NT$12,000 0.2% 4.47% NT$-2,000 Profile 6 Oversea Less 5 year Bond=3.75% Bond=100% 3.75% (NT$375) NT$10,375 0.0146% 1.21% NT$ 0 Profile 10 Domestic Less 5 year Bond=2% Risk-Free Saving=3% Bond=50% Risk-Free Saving=50% 2.5% (NT$250) NT$10,250 0.014% 1.186% NT$ -50 Profile 14 Domestic Less 5 year Stock=20% Stock=100% 20% (NT$2,000) NT$12,000 0.2% 4.47% NT$ -2,000 Profile 3 Domestic Less 1 year Bond=2% Stock=20% Bond=50% Stock=50% 11.1% ( NT$1,110) NT$11,110 0.0013% 0.37% NT$-1,050 Profile 7 Oversea Less 1 year Bond=3.75% Stock=14% Stock=50% Bond=50% 8.88% (NT$888) NT$10,888 0.00222% 0.4719% NT$ -700 Profile 11 Domestic Less 5 year Bond=20% Bond=100% 20% (NT$2,000) NT$12,000 0.2% 4.47% NT$ -2,000 Profile 15 Domestic Less 5 year Risk-Free Saving=2% Risk-Free Saving=100 % 2% (NT$200) NT$10,200 0.0128% 1.13% NT$ 0 Profile 4 Domestic Less 1 year Bond=20% Bond=100% 20% (NT$2,000) NT$12,000 0.2% 4.47% NT$-2,000 Profile 8 Oversea Less 1 year Bond=3.75% Risk-Free Saving=2.5% Bond=50% Risk-Free Saving=50% 3.12% (NT$313) NT$10,313 0.027% 1.632% NT$ -63 Profile 12 Domestic Less 5 year Risk-Free Saving=3% Risk-Free Saving=100% 3% (NT$300) NT$10,300 0.0146% 1.21% NT$ 0 Profile 16 Oversea Less 1 year Risk-Free Saving=14% Risk-Free Saving=100% 14% (NT$1,400) 11,400 0.56% 7.48% NT$ 0

31

Table 5 Descriptive Statistics of Sample


Category Socio-Economic Status Gender Male Female Total Responses Age (Years Old) 21~30 31~40 41~50 51~70 Total Responses Profession Finance Government Service Self-Employment Service Manufacturing Student Unemployment Others Total Responses Education Level Less than Junior High School Senior High school University Master or Ph.D. Total Responses Number of children and the Elder Non At Lest one Person Total Responses Material Status Married Single Others Total Responses 2,052 1,332 36 3,420 60 38.95 1.05 100 108 516 1,872 924 3,420 3.16 15.09 54.74 27.02 100 516 768 156 624 144 900 84 228 3,420 15.09 4.56 18.25 4.21 26.32 2.46 6.67 100 1,572 852 576 420 3,420 45.96 24.91 16.84 12.28 100 1,728 1,692 3,420 50.53 49.47 100 Observations % Category Socio-Economic Status Personal Labor Income per Month (NT$) NT$10,000<Income<NT$20,000 NT$20,001<Income<NT$30,000 NT$30,001<Income<NT$40,000 NT$40,001<Income<NT$50,000 Income>NT$50,001 Total Responses Household Income Source Independent Single salary Double salaries Total Responses Hedging Accumulate Wealth Trading Earning Higher Returns Diversifying Risk Preferring Fixed Incomes Total Responses Information Sources of Investment The Relatives or Friends Television or Radio Internet Financial Advisors News Paper or Magazines Total Responses 2,460 960 3,420 71.93 28.07 100 924 1,392 731 252 120 3,420 27.03 40.71 21.38 7.37 3.51 100 22.46 Investment Goals 936 1,272 756 348 60 48 3,420 27.37 37.19 22.11 10.18 1.75 1.4 100 2,076 156 1,188 3,420 60.7 4.56 34.74 100 1,080 480 744 684 432 3,420 31.58 14.04 21.75 20 12.63 100 Observations %

32

Table 6 Socio-Economic Determinants of Financial Risk Tolerance


Empirical specification: Model (1): (Risk Tolerance)=0+1(Female)+2(Age)+3(Age)2+4(Number of Children and the Elder)+5(Married) +6~9(Education)+10(Personal Labor Income)+11(Household Income)+ Model (2): (Risk Tolerance)=0+1(Female)+2(Age)+3(Age)2+4(Number of Children and the Elder)+5(Married) +6~9(Education)+10~13(Personal Labor Income)+14~18(Household Income)+ Independent Variables Constant Female Age Age2 Number of Children or the Elder Married Education Level Less than Junior High School Senior High School University Master or Ph.D. Personal Labor Income Level (NT$) Level 1:NT$21,000~NT$30000 Level 2:NT$31,000~NT$40,000 Level 3:NT$41,000~NT$50,000 Level 4: More than NT$50,000 Household Income Level ( NT$) Level 1: NT$20,000~NT$40,000 Level 2: NT$40,000~NT$60,000 Level 3: NT $60,000~NT$80,000 Level 4: NT $80,000~NT$100,000 Level 5: More than NT$100,000 Observations R2 Adjusted R2 F Statistics 3,420 0.110 0.108 52.79*** Wald Tests of Coefficient Equality F Statistics Personal Income: Level 2= Level 3= Level 4= Level 5 Level 4= Level 5 Family Income: Level 3= Level 4= Level 5 Education Level: Level 1= Level 3= Level 4= Level 5 10.88*** 0.000 Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively. 37.54*** 0.000 2.430*** 3.280*** 0.064 0.070 P-value 1.444*** (6.673) -7.642*** 3.041* -1.569 3.668** 4.858*** 3,420 0.149 0.145 33.08*** (-4.411) (1.746) (-0.919) (2.168) (2.709) 0.742*** (5.497) 1.356*** 2.176*** 2.660*** 1.580** (2.757) (4.640) (5.198) (2.512) Dependent Variable: Overall Scores of Risk Tolerance Model (1) Model (2) Coefficients (t-statistics) Coefficients (t-statistics) 17.291*** 1.042*** 0.047 0.179 -0.761* -5.081*** -0.302 (14.66) (7.472) (0.0507) (0.962) (-1.876) (-4.785) (-1.422) -0.329 4.059*** 3.314** 1.941 (-0.184) (2.909) (2.468) (1.434) 13.930*** 0.979*** 0.208 0.040 -0.612 -0.048 (9.096) (6.456) (0.218) (0.203) (-1.398) (-0.0291)

33

Table 7 Socio-Economic Determinants of Financial Risk Perceptions


Empirical specification: Model (1): (Risk Perceptions)=0+1(Female)+2(Age)+3(Age)2+4(Number of Children and the Elder) +5(Married) +6~9(Education)+10(Personal Labor Income)+11(Household Income)+ Model (2): (Risk Perceptions)=0+1(Female)+2(Age)+3(Age)2+4(Number of Children and the Elder) +5(Married) +6~9(Education)+10~13(Personal Labor Income)+14~18(Household Income)+ Dependent Variable: Overall Scores of Risk Perceptions Independent Variables Constant Female Age Age
2

Model (1) Coefficients 27.757*** -0.297 -2.612** 0.663** -0.334 -4.726*** 0.835*** (t-statistics) (16.86) (-1.527) (-2.041) (2.552) (-0.591) (-3.190) (2.816) 0.609 1.679 3.097

Model (2) Coefficients 29.149*** -0.405* -2.653** 0.703** 0.481 -2.897 (t-statistics) (13.43) (-1.885) (-1.961) (2.485) (0.776) (-1.233) (0.240) (0.849) (1.628) (1.798) (4.337) (1.647) (3.567) (6.140) (-1.052) (0.811) (-0.248) (1.641) (0.931) 3,420 0.047 0.042 9.242*** P value 0.000 0.001 0.000 1.0107

Number of Children and the Elder Married Education Level Less Than Junior High School Senior High School University Master or Ph.D. Personal Labor Income Level (NT$) Level 1:NT$21,000~NT$30000 Level 2:NT$31,000~NT$40,000 Level 3:NT$41,000~NT$50,000 Level 4: More than NT$50,000 Household Income Level (NT$) Level 1: NT$20,000~NT$40,000 Level 2: NT$40,000~NT$60,000 Level 3: NT $60,000~NT$80,000 Level 4: NT $80,000~NT$100,000 Level 5: More than NT$100,000 Observations R2 Adjusted R2 F Statistics

3.447* 1.041*** (5.528) 3.021*** 1.094* 2.586*** 5.470*** 0.904*** (2.995) -2.582 2.002 -0.601 3.932 2.366 3,420 0.033 0.031 14.650*** Wald Tests of Coefficient Equality F statistics

Personal Income: Level 2= Level 3= Level 4= Level 5 Level 4= Level 5 Family Income: Level 3= Level 4= Level 5 Education Level: Level 1= Level 3= Level 4= Level 5 2.070*** Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively. 12.270*** 9.630*** 11.640***

34

Table 8 Individual Investors Preference for Investment Attributes on Portfolio Choices: Controlling for Effects of Risk Tolerance and Risk Perception
Empirical specification: Model (1): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+ Model (2): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk T oleran ce + Model (3): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk T oleran ce +15 R isk T oleran ce + Model (4): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk Perception s + Model (5): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk Perception s +16 R isk Perception s + Model (6): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk T oleran ce +16 R isk Perception s + Model (7): (Portfolio)=0+1(International Portfolio)+2(Time Horizons)+3~5(Return on Investment)+6~8(Investment Weight)+9~10(Total Return) +11(Expected Value)+12(Variance)+13(Standard Deviation)+14(Expected Loss)+15 R isk T oleran ce +16 R isk Perception s +17 R isk Perception s +18 R isk Perception s

) +
2

Investment Attributes International Portfolio Investment Time Horizons Return on Bond Investment Return on Stock Investment Return on Risk-Free Saving Weight on Stock Investment Weight on Bond Investment Weight on Risk-Free Saving Total Return of Portfolio in Percent (%) Total Return of Portfolio in Amount (NT$) Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio

Basic Model (1) -0.471** (-2.346) -0.511*** (-2.824) -0.168*** (-4.421) -0.020 (-0.957) -0.178*** (-4.350) -0.001 (-0.140) 0.026*** (4.345) 0.029*** (5.151) 7.080 (1.317) -0.069 (-1.282) -0.000* (-1.954) -0.416 (-0.515) -0.089 (-1.270) -0.000*** (-2.946)

R isk T oleran ce

(R isk (R isk

T olerance)

Risk Tolerance Model (2) Model (3) -0.472** -0.496** (-2.349) (-2.463) -0.518*** -0.529*** (-2.866) (-2.926) -0.173*** -0.177*** (-4.545) (-4.623) -0.020 -0.021 (-0.980) (-1.000) -0.179*** -0.182*** (-4.385) (-4.477) -0.001 -0.002 (-0.321) (-0.488) 0.025*** 0.025*** (4.290) (4.226) 0.028*** 0.028*** (5.023) (4.953) 7.019 7.121 (1.313) (1.335) -0.068 -0.069 (-1.278) (-1.298) -0.000** -0.000** (-2.018) (-1.998) -0.474 -0.517 (-0.585) (-0.636) -0.084 -0.083 (-1.198) (-1.177) -0.000*** -0.000*** (-2.853) (-2.785) 0.016* 0.136*** (1.864) (3.734) -0.005*** (-3.483)

Risk Perception Model (4) Model (5) -0.477** -0.481** (-2.374) (-2.391) -0.520*** -0.490*** (-2.875) (-2.694) -0.175*** -0.170*** (-4.604) (-4.439) -0.020 -0.018 (-0.982) (-0.873) -0.179*** -0.176*** (-4.408) (-4.328) -0.002 -0.002 (-0.437) (-0.430) 0.025*** 0.025*** (4.239) (4.162) 0.028*** 0.028*** (4.951) (4.940) 7.024 7.057 (1.318) (1.325) -0.068 -0.069 (-1.282) (-1.290) -0.000** -0.000** (-2.035) (-2.005) -0.511 -0.536 (-0.629) (-0.659) -0.081 -0.078 (-1.157) (-1.101) -0.000*** -0.000*** (-2.795) (-2.790)

R isk Perceptions
Perceptions )
2

0.014*** (2.777)

Observations 3,411 3,411 3,411 3,411 Log Likelihood -1,148 -1,146 -1,140 -1,143 202.1*** 205.8*** 217.8*** 210.3*** 2 Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively.

-0.052 (-1.353) 0.002* (1.712) 3,411 -1,142 213.3***

Combination Model (6) Model (7) -0.498** -0.504** (-2.472) (-2.500) -0.504*** -0.482*** (-2.776) (-2.645) -0.171*** -0.167*** (-4.484) (-4.369) -0.019 -0.017 (-0.902) (-0.819) -0.180*** -0.178*** (-4.408) (-4.365) -0.002 -0.002 (-0.536) (-0.529) 0.024*** 0.024*** (4.099) (4.062) 0.028*** 0.028*** (4.884) (4.900) 7.169 7.202 (1.345) (1.352) -0.070 -0.070 (-1.309) (-1.317) -0.000* -0.000* (-1.931) (-1.936) -0.542 -0.575 (-0.667) (-0.705) -0.079 -0.075 (-1.119) (-1.064) -0.000*** -0.000*** (-2.770) (-2.776) -0.116*** -0.250 (-3.502) (1.265) -0.008* (-1.830) 0.081*** -0.170 (3.953) (-1.326) 0.004* (1.914) 3,411 3,411 -1,137 -1,135 226.4*** 222.6****

35

Table 9 Impacts of Risk Tolerance, Risk Perception and Socio-Economic Status on Individual Investors Portfolio Choices: All Correspondents
Independent Variables All Correspondents Risk Tolerance Risk Perceptions
Coefficients (t-statistics) Coefficients (t-statistics)

Portfolio Attributes International Portfolio -0.565*** (-2.606) -0.543** (-2.511) Investment Time Horizons -0.653*** (-3.344) -0.622*** (-3.173) Return of Bond Investment -0.200*** (-5.019) -0.193*** (-4.851) Return of Stock Investment -0.029 (-1.349) -0.027 (-1.240) -0.203*** (-4.759) -0.198*** (-4.664) Return of Risk-Free Saving Weight of Stock Investment -0.004 (-0.766) -0.003 (-0.646) Weight of Bond Investment 0.061*** (4.339) 0.068*** (4.868) Weight of Save Accounting 0.028*** (4.797) 0.028*** (4.811) Total Return of Portfolio in Percent (%) 9.473* (1.699) 9.344* (1.688) Total Return of Portfolio in Amount (NT$) -0.092* (-1.660) -0.091* (-1.650) Total Expected Value of Portfolio 0.000 (-1.625) -0.000* (-1.654) Total Variance of Portfolio -0.710 (-0.828) -0.726 (-0.845) Total Standard Deviation of Portfolio -0.084 (-1.127) -0.077 (-1.034) Total Expected Loss of Portfolio -0.000*** (-3.013) -0.000*** (-3.072) Wealth Management Advising 0.099 (1.407) 0.114 (1.617) Socio-Economic Status Age (Level) -0.002 (-1.049) -0.002 (-1.358) Education (Level) -0.002 (-1.129) -0.002 (-1.509) Personal Labor Income (Level) -0.002* (-1.940) -0.002** (-2.001) -0.005** (-2.169) -0.006*** (-2.681) Gender Occupation (Level) 0.000 (-0.030) 0.000 (-0.149) 0.002 (0.652) 0.002 (0.819) Number of Children and the Elder Marital Status -0.013*** (-3.357) -0.014*** (-3.381) Family Income Level 0.002** (2.054) 0.002** (2.119) Financial Related Background Financial Professional Training -0.007** (-2.265) -0.006* (-1.903) Certificate of Financial Management -0.007* (-1.848) -0.007* (-1.943) Major in Degree of Finance -0.005* (-1.654) -0.004 (-1.379) Information Sources of Investment The Relatives 0.000 (0.128) 0.000 (0.063) Mass Media 0.001 (0.405) 0.002 (0.676) Internet 0.007*** (2.587) 0.006** (2.326) 0.019*** (4.417) 0.017*** (3.857) Financial Advisor News Paper or Magazines 0.005* (1.856) 0.004 (1.379) Investment Goals Hedging 0.006 (1.537) 0.006* (1.696) Increase Wealth -0.006** (-2.104) -0.005** (-1.990) Trading -0.007* (-1.927) -0.007* (-1.759) Earning Higher Return 0.010*** (3.672) 0.010*** (3.514) -0.003 (-1.118) -0.002 (-0.581) Diversifying Risk Preferring Fixed Income -0.002 (-0.602) -0.001 (-0.265) Risk Measures Standard Deviation -0.003 (-1.571) -0.002 (-1.358) Expected of Loss 0.003* (1.677) 0.003 (1.584) Expected Value of Loss -0.001 (-0.588) -0.001 (-0.732) Semi-Variance -0.006** (-2.482) -0.007*** (-2.937) Risk Tolerance Overall Scores on Risk Tolerance 0.161*** (4.163) (Overall Scores on Risk Tolerance)2 -0.006*** (-3.979) Risk Perceptions Overall Scores on Risk Perceptions -0.067* (-1.660) 0.002** (2.037) (Overall Scores on Risk Perceptions)2 Observations 3,411 3,411 Log Likelihood -1,072 -1,076 352.6*** 346.2*** 2 Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively.

36

Table 10 Impacts of Risk Tolerance, Risk Perceptions and Socio-Economic Status on Portfolio Choices: Male versus Female Investors
Independent Variables Male Investors Risk Tolerance Risk Perceptions
Coefficients (t-statistics) Coefficients (t-statistics)

Female Investors Risk Tolerance Risk Perceptions


Coefficients (t-statistics) Coefficients (t-statistics)

Portfolio Attributes International Investment -0.754** (-2.128) -0.536 (-1.486) Investment Duration -0.40 (-1.211) -0.432 (-1.270) Return of Bond Investment -0.209*** (-3.189) -0.198*** (-2.943) Return of Stock Investment -0.028 (-0.758) -0.028 (-0.747) Return of Risk-Free Saving -0.243*** (-3.553) -0.233*** (-3.313) Weight of Stock Investment -0.020 (-1.457) -0.024 (-1.585) Weight of Bond Investment 0.003 (0.109) -0.000 (-0.009) Weight of Save Accounting 0.018 (1.240) 0.011 (0.714) Total Return of Portfolio in Percent 20.451** (2.001) 22.578** (2.153) Total Return of Portfolio in Amount -0.201** (-1.971) -0.223** (-2.127) Total Expected Value of Portfolio 0.000 (0.395) 0.000 (0.767) Total Variance of Portfolio 2.630* (1.905) 3.092** (2.166) Total Standard Deviation of Portfolio -0.391*** (-3.082) -0.402*** (-3.055) Total Expected Loss of Portfolio -0.001*** (-4.432) -0.001*** (-4.510) Wealth Management Advising -0.009 (-0.085) -0.034 (-0.307) Socio-Economic Status Age Level 0.003 (0.905) 0.001 (0.437) Education Level 0.008* (1.847) 0.005 (1.175) Income Level -0.002 (-1.005) -0.002 (-0.883) Occupation -0.002* (-1.689) -0.002 (-1.299) Number of Independent Persons 0.011 (1.644) 0.015** (2.202) Marital status -0.012* (-1.743) -0.015* (-1.947) Family income -0.001 (-0.550) -0.000 (-0.173) Financial Related Background Financial Professional Training -0.011 (-1.616) -0.009 (-1.309) Certificate of Financial Management -0.014* (-1.806) -0.013 (-1.615) Major in Finance -0.012* (-1.747) -0.009 (-1.272) Information Sources of Investment The Relatives 0.003 (0.510) 0.004 (0.699) Mass Media -0.005 (-1.039) -0.004 (-0.820) Internet 0.014** (2.188) 0.014** (2.097) Financial Advisor 0.029*** (3.647) 0.021*** (2.676) News Paper or Magazines 0.001 (0.275) -0.001 (-0.222) Investment Goal Hedging 0.016* (1.805) 0.022** (2.458) Increase Wealth -0.011** (-2.131) -0.008 (-1.550) Trading -0.017** (-2.268) -0.016** (-2.186) Earning Higher Return 0.023*** (3.796) 0.026*** (4.051) Diversifying risk -0.006 (-1.141) -0.005 (-0.995) Preferring Fixed Income -0.005 (-1.124) -0.002 (-0.426) Risk Measures Standard Deviation -0.015*** (-4.383) -0.009** (-2.565) Expected of Loss 0.007** (1.978) 0.010*** (2.687) Expected Value of Loss 0.003 (0.874) 0.002 (0.502) Semi-Variance 0.001 (0.112) -0.005 (-1.044) Risk Tolerance Overall Scores on Risk Tolerance 0.326*** (3.644) (Overall Scores on Risk Tolerance)2 -0.013*** (-3.425) Risk Perceptions Overall Scores on Risk Perceptions -1.045*** (-5.157) (Overall Scores on Risk Perceptions)2 0.035*** (5.114) Observations 1,728 1,728 Log Likelihood -455.2 -430.600 2 355.3*** 404.4*** Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively.

-0.069 -1.075*** -0.218*** -0.047 -0.165** 0.007 0.059*** 0.031*** 11.686 -0.116 -0.000** -2.871** 0.138 0.000 0.138 -0.003 -0.002 -0.004** 0.001 -0.004 -0.026*** 0.007*** -0.002 -0.008 0.004 0.006 0.004 0.004 0.021*** 0.007 0.004 0.005 -0.008 0.001 0.005 -0.004 0.004 -0.002 -0.004 -0.009** 0.101** -0.004*

(-0.209) (-3.622) (-3.700) (-1.419) (-2.551) (1.043) (2.850) (3.873) (1.425) (-1.411) (-2.175) (-2.109) (1.149) (0.671) (1.344) (-0.926) (-0.476) (-2.024) (0.936) (-0.822) (-2.630) (2.936) (-0.362) (-1.377) (1.041) (1.564) (1.054) (0.981) (3.076) (1.598) (0.795) (1.007) (-1.304) (0.225) (1.081) (-1.026) (1.306) (-0.685) (-1.421) (-2.490) (2.034) (-1.754)

-0.043 -1.099*** -0.220*** -0.049 -0.163** 0.007 0.065*** 0.032*** 11.537 -0.114 -0.000** -2.769** 0.130 0.000 0.141 -0.004 -0.002 -0.004* 0.001 -0.004 -0.025*** 0.007*** -0.001 -0.009 0.005 0.007 0.004 0.004 0.019*** 0.006 0.004 0.004 -0.007 0.001 0.006 -0.004 0.003 -0.002 -0.004 -0.009**

(-0.132) (-3.694) (-3.732) (-1.485) (-2.526) (1.157) (3.228) (3.910) (1.405) (-1.391) (-2.196) (-2.043) (1.088) (0.652) (1.374) (-1.251) (-0.733) (-1.930) (0.933) (-0.773) (-2.614) (2.994) (-0.236) (-1.488) (1.170) (1.634) (1.035) (1.018) (2.871) (1.542) (0.794) (0.894) (-1.176) (0.157) (1.236) (-1.077) (1.012) (-0.590) (-1.519) (-2.524)

0.037 -0.001 1,683 -523.8 184.2***

(0.638) (-0.473)

1,683 -525.300 181.2***

37

Table 11 Impacts of Risk Tolerance, Risk Perceptions and Socio-Economic Status on Portfolio Choices: Investors with Low versus High Income
Independent Variables Investors with Low Income Risk Tolerance Risk Tolerance
Coefficients (t-statistics) Coefficients (t-statistics)

Investors with High Income Risk Tolerance Risk Perceptions


Coefficients (t-statistics) Coefficients (t-statistics)

Portfolio Attributes International Investment -0.824** (-2.366) -0.908** (-2.553) Investment Duration -0.162 (-0.514) -0.040 (-0.123) Return of Bond Investment -0.146** (-2.356) -0.128** (-2.020) Return of Stock Investment 0.017 (0.485) 0.029 (0.796) Return of Risk-Free Saving -0.144** (-2.058) -0.146** (-2.055) Weight of Stock Investment -0.012* (-1.695) -0.015** (-2.125) Weight of Bond Investment 0.014 (0.544) 0.007 (0.259) Weight of Save Accounting 0.026*** (3.136) 0.024*** (2.868) Total Return of Portfolio in Percent 5.514 (0.645) 5.566 (0.654) Total Return of Portfolio in Amount -0.053 (-0.621) -0.054 (-0.632) Total Expected Value of Portfolio -0.000 (-1.624) -0.000 (-1.286) Total Variance of Portfolio -1.891 (-1.325) -2.268 (-1.568) Total Standard Deviation of Portfolio 0.020 (0.155) 0.052 (0.406) Total Expected Loss of Portfolio -0.000 (-1.526) -0.000 (-1.357) Wealth Management Advising 0.024 (0.214) 0.019 (0.168) Socio-Economic Status Age Level -0.010* (-1.890) -0.010* (-1.918) Education Level -0.000 (-0.134) -0.001 (-0.430) Gender -0.004 (-0.797) -0.008 (-1.464) Occupation 0.005*** (3.043) 0.006*** (3.484) Number of Independent Persons 0.007 (0.556) 0.011 (0.886) Marital status 0.003 (0.378) -0.003 (-0.304) Family income 0.004 (0.886) 0.006 (1.234) Financial related background Financial Professional Training -0.022*** (-4.090) -0.020*** (-3.609) Certificate of Financial Management -0.005 (-0.931) -0.007 (-1.224) Major in Finance -0.010** (-1.988) -0.008 (-1.571) Sources of Investment Information The Relatives -0.005 (-0.880) -0.009 (-1.523) Mass Media -0.006 (-1.291) -0.008 (-1.520) Internet 0.013** (2.169) 0.011* (1.933) Financial Advisor 0.009 (0.863) 0.009 (0.787) News Paper or Magazines 0.002 (0.340) -0.003 (-0.611) Investment Goal Hedging 0.028*** (3.036) 0.027*** (2.980) Increase Wealth -0.011** (-2.240) -0.011** (-2.305) Trading -0.008 (-0.873) -0.006 (-0.668) Earning Higher Return 0.005 (0.888) 0.008 (1.329) Diversifying risk -0.004 (-0.670) -0.001 (-0.203) Preferring Fixed Income 0.011** (2.264) 0.013** (2.515) Risk measures Standard Deviation 0.000 (0.136) 0.005 (1.550) Expected of Loss 0.010** (2.541) 0.012*** (2.885) Expected Value of Loss -0.003 (-0.662) -0.003 (-0.684) Semi-Variance -0.001 (-0.125) -0.002 (-0.370) Risk Tolerance Overall Scores on Risk Tolerance 0.104 (1.532) (Overall Scores on Risk Tolerance)2 -0.003 (-1.028) Risk Perceptions Overall Scores on Risk Perceptions -0.297*** (-3.798) (Overall Scores on Risk Perceptions)2 0.009*** (4.037) Observations 1,551 1,551 Log Likelihood -462.6 -449.0 2 209.7*** 237.1*** Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively.

-0.715** -0.730*** -0.226*** -0.050* -0.238*** -0.009 0.101*** 0.023** 18.034** -0.178** 0.000 -0.774 -0.137 -0.000* 0.193** -0.002 -0.010*** -0.010*** -0.001 -0.002 -0.022*** 0.004** 0.004 -0.016** 0.003 0.005 0.004 0.012** 0.022*** 0.002 0.009* 0.000 -0.011** 0.002 -0.004 -0.013*** 0.001 -0.002 -0.003 -0.006* 0.201*** -0.008***

(-2.238) (-2.648) (-4.021) (-1.650) (-4.003) (-1.012) (4.160) (2.198) (2.096) (-2.071) (0.240) (-0.657) (-1.336) (-1.856) (1.977) (-1.028) (-3.264) (-2.740) (-0.909) (-0.526) (-2.905) (2.573) (0.567) (-2.488) (0.774) (1.295) (0.904) (2.391) (3.717) (0.479) (1.722) (0.0855) (-1.989) (0.529) (-1.134) (-3.234) (0.340) (-0.461) (-1.015) (-1.865) (3.635) (-3.721)

-0.663** -0.796*** -0.236*** -0.054* -0.239*** -0.008 0.113*** 0.024** 17.258** -0.170** 0.000 -0.679 -0.143 -0.000** 0.203** -0.003 -0.011*** -0.012*** -0.001 -0.002 -0.022*** 0.004** 0.005 -0.016** 0.004 0.005 0.004 0.011** 0.018*** 0.001 0.009* -0.000 -0.010* 0.002 -0.003 -0.013*** 0.000 -0.001 -0.004 -0.007**

(-2.086) (-2.885) (-4.182) (-1.793) (-4.025) (-0.854) (4.732) (2.274) (2.034) (-2.008) (0.146) (-0.579) (-1.399) (-2.050) (2.082) (-1.399) (-3.399) (-3.218) (-1.250) (-0.488) (-2.807) (2.560) (0.748) (-2.540) (0.933) (1.324) (1.039) (2.327) (3.097) (0.130) (1.758) (-0.0875) (-1.818) (0.466) (-0.830) (-3.202) (0.0615) (-0.268) (-1.378) (-2.103)

0.090 -0.002 1,860 -560.2 241.8***

(1.312) (-1.167)

1,860 -566.5 229.4***

38

Table 12 The Impact of Risk Tolerance and Socio-Economic Status on Portfolio Choices: Low Education versus High Education
Independent Variables Investors with Low Education Risk Tolerance Risk Perceptions
Coefficients (t-statistics) Coefficients (t-statistics)

Investors with High Education Risk Tolerance Risk Perceptions


Coefficients (t-statistics) Coefficients (t-statistics)

Portfolio Attributes International Investment -1.202** (-2.419) -1.145** (-2.299) Investment Duration 0.018 (0.041) 0.053 (0.117) Return of Bond Investment -0.152** (-2.005) -0.141* (-1.874) Return of Stock Investment -0.030 (-0.662) -0.027 (-0.591) Return of Risk-Free Saving -0.149* (-1.763) -0.140* (-1.661) Weight of Stock Investment -0.003 (-0.342) -0.001 (-0.123) Weight of Bond Investment -0.093 (-0.449) -0.136 (-0.653) Weight of Save Accounting 0.030*** (2.884) 0.031*** (2.948) Total Return of Portfolio in Percent 33.035 (1.337) 33.669 (1.315) Total Return of Portfolio in Amount -0.329 (-1.330) -0.335 (-1.309) Total Expected Value of Portfolio 0.000 (0.188) 0.000 (0.124) Total Variance of Portfolio -3.094 (-1.483) -3.343 (-1.593) Total Standard Deviation of Portfolio -0.059 (-0.389) -0.029 (-0.189) Total Expected Loss of Portfolio 0.000 (0.240) 0.000 (0.313) Wealth Management Advising 0.000 (0.002) 0.024 (0.111) Socio-Economic Status Age Level 0.032 (0.625) 0.042 (0.811) Income Level 0.003 (0.752) 0.002 (0.613) Gender 0.020 (0.830) 0.022 (0.901) Occupation -0.003 (-0.724) -0.002 (-0.489) Number of Independent Persons 0.011 (0.721) 0.015 (0.962) Marital status -0.027 (-0.642) -0.037 (-0.849) Family income Level -0.005 (-0.569) -0.006 (-0.699) Financial related background Financial Professional Training -0.445*** (-8.787) -0.469 (-0.144) Certificate of Financial Management -0.039 (-0.417) -0.058 (-0.625) Major in Finance -0.045 (-0.796) -0.049 (-0.880) Information Sources of Investment The Relatives 0.002 (0.197) 0.001 (0.0849) Mass Media 0.007 (0.712) 0.010 (0.960) Internet 0.061 (0.758) 0.077 (0.964) Financial Advisor 0.116 (0.639) 0.150 (0.821) News Paper or Magazines 0.088 (0.669) 0.116 (0.874) Investment Goal Hedging -0.067 (-0.641) -0.085 (-0.816) Increase Wealth 0.005 (0.471) 0.002 (0.150) Trading 0.403 () 0.429*** (8.694) Earning Higher Return -0.023 (-0.791) -0.032 (-1.056) Diversifying risk -0.001 (-0.105) -0.002 (-0.157) Preferring Fixed Income -0.085 (-0.726) -0.110 (-0.934) Risk Measures Standard Deviation -0.009 (-0.517) -0.012 (-0.652) Expected of Loss -0.008 (-0.794) -0.013 (-1.171) Expected Value of Loss 0.048 (0.739) 0.063 (0.969) Semi-Variance -0.003 (-0.223) 0.002 (0.121) Risk Tolerance Overall Scores on Risk Tolerance 0.003 (0.0834) (Overall Scores on Risk Tolerance)2 -0.000 (-0.344) Risk Perceptions Overall Scores on Risk Perceptions -0.048 (-1.545) (Overall Scores on Risk Perceptions)2 0.001* (1.711) Observations 624 624 Log Likelihood -342.0 -340.9 2 110.3*** 112.6*** Note: *, **, *** are denoted statistically significant at level of 10%, 5%, and 1%, respectively.

-0.400 -0.768*** -0.221*** -0.021 -0.221*** -0.004 0.054*** 0.032*** 5.442 -0.052 -0.000** -0.857 -0.028 -0.000*** 0.122 -0.003 -0.001 -0.006** 0.000 0.002 -0.011** 0.001 -0.006* -0.007* -0.004 -0.002 0.003 0.006* 0.018*** 0.005 0.002 -0.005* -0.009** 0.015*** -0.003 -0.003 -0.002 0.005** -0.002 -0.005* 0.228*** -0.009***

(-1.579) (-3.460) (-4.721) (-0.847) (-4.429) (-0.656) (3.783) (4.541) (0.918) (-0.877) (-2.302) (-0.895) (-0.340) (-3.252) (1.516) (-1.341) (-0.878) (-1.983) (0.345) (0.491) (-2.136) (1.149) (-1.730) (-1.807) (-1.200) (-0.613) (0.931) (1.801) (3.641) (1.540) (0.557) (-1.813) (-2.060) (4.456) (-1.046) (-1.055) (-1.095) (2.265) (-1.156) (-1.724) (4.893) (-4.733)

-0.366 -0.783*** -0.220*** -0.022 -0.221*** -0.003 0.063*** 0.032*** 4.963 -0.047 -0.000** -0.808 -0.027 -0.000*** 0.144* -0.004* -0.001 -0.007*** 0.000 0.002 -0.011** 0.001 -0.004 -0.007* -0.003 -0.002 0.004 0.004 0.013*** 0.003 0.003 -0.004 -0.007* 0.014*** -0.001 -0.003 -0.002 0.005** -0.003 -0.008***

(-1.451) (-3.535) (-4.716) (-0.877) (-4.448) (-0.502) (4.477) (4.578) (0.846) (-0.805) (-2.334) (-0.846) (-0.325) (-3.416) (1.802) (-1.807) (-0.808) (-2.663) (0.0315) (0.480) (-2.021) (1.084) (-1.190) (-1.836) (-1.100) (-0.697) (1.290) (1.320) (2.765) (0.863) (0.671) (-1.505) (-1.796) (4.276) (-0.318) (-0.925) (-0.959) (2.262) (-1.220) (-2.588)

2,787 -861.8 -316.8***

-0.062 0.002* 2,787 -868.4 303.5***

(-1.397) (1.852)

39

Table 13 Summary Results of the Impact of Risk Tolerance on Individual Portfolio Choices
Category Portfolio Attributes International Investment Investment Duration Return of Bond Investment Return of Stock Investment Return of Risk-Free Saving Weight of Stock Investment Weight of Bond Investment Weight of Save Accounting Total Return of Portfolio in Percent (%) Total Return of Portfolio in Amount (NT$) Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Wealth Management Advising Sources of Investment Information The Relatives Mass Media Internet Financial Advisor News Paper or Magazines Investment Goal Hedging Increasing Wealth Trading Earning Higher Return Diversifying Risk Preferring Fixed Income Financial Related Background Financial Professional Training Certificate of Financial Management Major in Finance Risk Measures Standard Deviation Expected Loss Expected Value of Loss Semi-Variance Risk Tolerance Overall Scores on Risk Tolerance Square of Overall Scores on Risk Tolerance All Correspondents Gender Female Male Income Low High Education Low High

40

Table 14 Summary Results of the Impact of Risk Perceptions on Individual Portfolio Choices
Category Portfolio Attributes International Investment Investment Duration Return of Bond Investment Return of Stock Investment Return of Risk-Free Saving Weight of Stock Investment Weight of Bond Investment Weight of Save Accounting Total Return of Portfolio in Percent (%) Total Return of Portfolio in Amount (NT$) Total Expected Value of Portfolio Total Variance of Portfolio Total Standard Deviation of Portfolio Total Expected Loss of Portfolio Wealth Management Advising Sources of Investment Information The Relatives Mass Media Internet Financial Advisor News Paper or Magazines Investment Goal Hedging Increasing Wealth Trading Earning Higher Return Diversifying Risk Preferring Fixed Income Financial Related Background Financial Professional Training Certificate of Financial Management Major in Finance Risk Measures Standard Deviation Expected Loss Expected Value of Loss Semi-Variance Risk Perceptions Overall Scores on Risk Perceptions Square of Overall Scores on Risk Perceptions All Correspondents Gender Female Male Income Low High Education Low High

41

Potrebbero piacerti anche