Nama: Fidelis Permana Sidi

Nim: 2018-0101-0004
Matkul: Statistika Bisnis 4
No. Urut: 6

6. Consider the ANOVA table that follows

Analysis of
variance
Source DF SS MS F
Regression 5 3710,00 742,00 12,89
Residual Error 46 2647,38 57,55
Total 51 6357,38

a. Determine the standard error of estimate. About 95% of the residual will be between
what two values?
2647.38
𝑠𝑦.12345 = √ = 7.586
52 − (5 + 1)
95% of the residuall will be between ±15.172, found by 2(7.586).
b. Determine the coefficient of multipe determination. Interpret this value.
3710.00
𝑅2 = = 0.583
6357.38
The independent variables explain 58.3% of the variation
c. Determine the coeffecient of multipe determinagtion, adjusted for the degrees of
freedom.
2647.38

𝑅 2 𝑎𝑑𝑗 = 1 − 46
6357.38 = 0.461
51
= 1 - 0.461 = 0.539

13. A multiple regression analysis yields the following partial result

Source Sum of Squares df
Regression 750 4
Error 500 55

a. What is the total sample size ?

n = 40
b. How many independent variables are being considered?
4
c. Compute the coefficient of determination.
750
𝑅2 = = 0,6
1.250
d. Compute the standard error of estimate.
𝑆𝑦.1234 = √500/35 =3,7796
e. Test the hypotesis that at least one of the regression coeffecients is not equal to zero.
Let α= ,05.
𝐻𝑜 : 𝛽1 = 𝛽2 = 𝛽3 = 𝛽4 = 0
 𝐻1 : 𝑛𝑜𝑡 𝑎𝑙𝑙 𝑡ℎ𝑒 𝛽𝑠 𝑒𝑞𝑢𝑎𝑙 𝑧𝑒𝑟𝑜
𝐻𝑂 𝑖𝑠 𝑟𝑒𝑗𝑒𝑐𝑡𝑒𝑑 𝑖𝑓 𝐹 > 2,65
750/4
F= 500/35 = 13,125
𝐻𝑜 𝑖𝑠 𝑟𝑒𝑗𝑒𝑐𝑡𝑒𝑑. 𝐴𝑡 𝑙𝑒𝑎𝑠𝑡 𝑜𝑛𝑒 𝛽𝑖 𝑑𝑜𝑒𝑠 𝑛𝑜𝑡 𝑒𝑞𝑢𝑎𝑙 𝑧𝑒𝑟𝑜.

27. An investment advisor is studying the realitionship between a common stock’s price to
earning (P/E) rasio and factors that she thinks would influence it. She has the following
data on the earnings per share and the dividend percentage (yield) for sample of 20 stocks.
a. Develope a multipe linear regression with P/E as the dependent variable.
ŷ = 29.913 − 5.324𝑥1 + 1.449𝑥2
b. Are either of the two independent variables an effective predictoŶŷr of P/E?
EPS is (t = -3.26, p-value = 0.005). yield is not (t = 0.81, p-value = 0.431).
c. Interpret the regression coefficients
An increase of 1 in EPS result in a decline of 5.324 in P/E.
d. Do any these stock look particulalry undervalued?
Stock number 2 is undervalued.
e. Plot the residuals and check the normally assumpption. Plot the fitted values againts
the residuals.
Below is a residual plot. It does not apper to follow the normal distribution.
f. Does there apper to be any problems with homoscedasticity?
There does not seem to be a problem with the plot of the residuals versus the fitted
values.

g. Develope a correlation matrix. Do any of the correlations indicate multicollinearity?

The correlation between yield and EPS is not a problem. No. Problem with
multicoliinearity.
P/E EPS
EPS -0.602
Yield .54 .162

20. Mike Wilde is president of the teachers’ union for Otsego School District. In preparing
for upcoming negotiations, he would like to investigate the salary structure of classroom
teachers in the district. He believes there are three factors that affect a teacher’s salary:
years of experience, a rating of teaching effectiveness given by the principal, and whether
the teacher has a master’s degree. A random sample of 20 teachers resulted in the
following data.

a. Develop a correlation matrix. Which independent variable has the strongest

correlation with the dependent variable? Does it appear there will be any problems
with multicollinearity?
Salary Years of Principals’s Master’s
experiance, 𝑥1 rating, 𝑥2 degree
Salary 1
Years of 0,577299 1
experiance, 𝑥1
Principals’s 0,442869 -0,32744 1
rating, 𝑥2
Master’s -0,37181 -0,80861 0,373321 1
degree
 Strongest Correlation:
=> Salary with Years of Experience
Problem with multicollineary:
Yes. Because there is -0.80861 (more than -0.7)

b. Determine the regression equation. What salary would you estimate for a teacher with
five years experience, a rating by the principal of 60, and no masters degree?

▸ Regression Equation:
Ŷ= 14,5768 + 1,5241𝑥1 + 0,1971𝑥𝑥 + 0,5135𝑥3
Ŷ= 14,5768 + 1,5241(5) + 0,1971(60) + 0,5135 (0)
Ŷ= 36,666
The estimated salary for a teacher with five years experience, a rating by the principal
of 60, and no master degree is 36,6658 (in thousand US$)

c. Conduct a global test hypothesis to determine whether any of the regression

coefficients differ from zero. Use the .05 significance level.

The critical value of F: numerator 3, denumerator 2: 19,2

Reject 𝐻0 if F > 19,2
𝑆𝑆𝑅/𝑘 𝑀𝑆𝑅
Using the anova table, the F-Statistics is F= 𝑆𝑆𝐸/[𝑛−(𝑘+1)] = = 2,385
𝑀𝑆𝐸
At least one coefficient is not zero.

d. Conduct a test of hypothesis for the individual regression coefficients. Would you
consider deleting any of the independent variables? Use the .05 significance level.

▸ Using the regression table, the t ratio of years experience is 1,5131 and the p values of
Years experience is 0.2694. Because the p value is more than 0.05, we conclude that
the years expereince regression coefficient could equal 0. Thus, years of experience
should not be included in the equation to predict a teacher’s salary.
▸ The t ratio of principal’s rating is 1,9658 and the p value is 0,1882. The p value is
more than 0.05 the principal rating could equal 0. So, principal’s rating should not be
included in the equation topredict teachers salary.
 ▸ The t ratio of master degree is 0,0952 and the p value is 0,9328. The p value is more
than 0.05 the principal rating could equal 0. So, principal’s rating should not be
included in the equation topredict teachers salary.
▸ Without Master degree

e. If your conclusion in part (d) was to delete one or more independent variables, run the
analysis again without those variables.
SUMMARY OUTPUT

Regression Statistics

Multiple R 0,883472994

R Square 0,78052453

Adjusted R Square 0,634207551

Standard Error 2,939992117

Observations 6

ANOVA

df SS MS F Significance F

Regression 2 92,21767238 46,10883619 5,334477 0,102820327

Residual 3 25,93066095 8,643553651

Total 5 118,1483333

Lower
Coefficients Standard Error t Stat P-value Lower 95% Upper 95% 95,0% Upper 95,0%

Intercept 15,00446535 6,017721046 2,493380007 0,08822 -4,146608764 34,1555395 -4,14660876 34,15553945

X Variable 1 1,449013691 0,512687599 2,826309227 0,066392 -0,182587063 3,08061445 -0,18258706 3,080614445

X Variable 2 0,198984871 0,080477956 2,472538834 0,089862 -0,057131902 0,45510165 -0,0571319 0,455101645

 f. Determine the resduals for the equation of part (E). Use steam and leaf chart
histogram to verify that the distributions of the residual is approximately normal.

100
80
60
40
20
0
Gaji (ribu $)
pengalaman mengajar (tahun),
penilaian kepala sekolah,

g. Does the plot reveal any violations of the as sumptions of regression?

50
40
Residu

30
20
10
0
0 10 20 30 40 50

Ŷ
31. a. Carry out a global test of hypothesis to verify if any of the regression coefficients are
different from zero.
The regression equation is
Auction price = -118,929 + 1.63 loan + 2.1 Monthly payments + 50 payments made
Analysis of
variance
Source DF SS MS F P
Regression 3 5966725061 1988908354 39.83 0.000
Residual 16 798944439 49934027
Error
Total 19 6765669500
The computed Fis 39.83 it is much large than the critical value 3.24. the p-value is also quite
small. Thus, the null hypotesis that all the regression coefficient are zero can be rejected. At
least one of the multipe regrresion coefficient is different from zero.
 b.
Predictor coef SE coef T P
Constant -118929 19734 -6.03 0.000
Loan 1.6268 0.1809 8.99 0.000
Monthly 2.06 14.95 0.14 0.892
Payment
Payments Made 50.3 234.9 0.37 0.714

The null hypotesis is than the coefficient is zero in the individual test. It would be rejected is t
is less than -2.120 or more than 2.120. in the case, the t value for the loan variable is large
than the critical value. Thus, it should not be removed. However, the monthly payment and
payments made variables would likely be removed.
c. the revised regression equation is: auction price = -119,893 + 1.67 loan