See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/259326743

Fractal Formation and Trend Trading Strategy in Futures Market

Conference Paper · December 2013

DOI: 10.1007/978-3-642-33914-1_38

CITATIONS READS

0 6,626

3 authors, including:

Saulius Masteika Audrius Lopata

Vilnius University Vilnius University
24 PUBLICATIONS   84 CITATIONS    37 PUBLICATIONS   165 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

High frequency trading algorithms in futures markets. Blockchain technology applications. View project

All content following this page was uploaded by Saulius Masteika on 02 February 2015.

The user has requested enhancement of the downloaded file.

FRACTAL FORMATION AND TREND
TRADING STRATEGY IN FUTURES
MARKET

Saulius Masteika1,2, Aleksandras V.Rutkauskas1, Audrius Lopata2,3

1
Department of Finance Engineering, Faculty of Business Management, VGTU, Vilnius,
Lithuania, EU, E-mail (corresponding author): saulius.masteika@vgtu.lt
2
Department of Informatics, Faculty of Humanities, Vilnius university, Lithuania,EU
3
Information Systems Department, Kaunas University of Technology, Lithuania, EU

Abstract The paper presents the details of trend trading algorithm in futures
market. A contribution of this paper lies in a modified chart pattern related to a
fractal formation, nonlinearity and chaos theory, broadly discussed by Benoit
B.Mandelbrot and Bill M. Williams. As typical fractal pattern often is being ap-
plied in conjunction with other forms of technical analysis, like moving averages,
Elliott Waves analysis or MACD indicators the proposed pattern is presented as a
basic indicator itself. The strategy can be applied as up-trend market forecasting
tool. The efficiency of the proposed strategy was tested with the most active North
American futures contracts using 10 years historical daily data. Experimental re-
sults showed better returns if compared to overall market average- CRB index.

1 Introduction
Last decade expansion of high frequency trading, quantitative analysis and au-
tomated trading shows an increasing belief that algorithmic techniques can be
helpful for decision making in financial markets [2, 4, 5]. Because of increase in
market volatility and growing demand for hedging businesses trading volumes
from stock and currency markets move to derivative markets [1]. The need for
trading algorithms and quantitative analysis in futures markets are in great de-
mand.

2 The basic concept of the proposed trading strategy

Despite the shortage of hard evidences about the profitability of trading algo-
rithms there are some research papers claiming that the application of momentum
trend trading and chart patterns can give some useful information to market partic-
ipants [3, 9, 10, 11]. The proposed trading strategy is up-trend following system
based on continuous bar chart formation.
 The proposed continuous chart formation is related to fractal pattern and chaos
theory, broadly discussed by Benoit B. Mandelbrot and Ph.D. Bill M.Williams [6,
11]. Decision to search for a new modified fractal was made because of inaccura-
cies when building back-testing trading strategies based on typical fractals. Inac-
curacies appear because of a big variety of possible formations, lack of signal
quality measurement and clear rules for closing positions. Typical fractal for-
mations are most often presented in conjunction with other forms of technical
analysis, like moving averages, Elliott Waves analysis or MACD indicator [11]. In
general, typical fractals can be as a good decision support tool, but not the basic
indicator itself. Decision to search for a more accurate, algorithmic, testable and
still conformable with trend trading and chaos theory chart pattern was made.

2.1 Chart Pattern for Trading Strategy

The proposed short term chart pattern is composed of only 3 consecutive chart
bars instead of five. The proposed pattern uses a modified truncated fractal to gen-
erate a buy and sell signals. A fractal occurs when there is a pattern with the high-
est high in the middle and one lower high on each side, as it can be seen in some
examples in Fig.1.

Fig. 1 Examples of the proposed chart pattern

The trading strategy opens a long position when the current price (i) tops the
previous bar’s (i-1) highest price. If current price bar (i) opens higher than (i-1)
and forms a gap, a buy signal is identified as an opening price. A trailing stop or-
der is set as the lowest level of the bar (i-1) price minus a tick size. The trailing
stop price is adjusted and moved to the next bar lowest price each time the new
bar is formed. If there is a sudden crash in the market and the next bar opens with
a gap down, sell price is set as an open price of this following bar.

2.2 Signal strength of the trading pattern

In order to avoid inaccuracies related to taking only a particular contract or the
most suitable time period for analysis the quality of signals and ranking techniques
of contracts must be considered [7]. The proposed trading strategy is looking for
futures contracts, which during two day’s period increased in a price the most, and
the lowest price of the last bar is higher than a month ago. The biggest increase in
a price of a particular contract during the period between i-4 and i-2 means the
best quality and the highest rank. If the chart pattern is formed with several con-
tracts at a time, the one with the highest rank is chosen for a trade.

3. Experimental setup
The efficiency of the strategy was tested with historical data from futures ex-
changes. The daily time series were collected from GLOBEX, NYMEX, ECBOT,
CFE, ICE and ICE-NYBOT exchanges. Test period from 2002 till 2012. Most ac-
tive futures were taken from these sectors: Energies, Metals, Grains, Financials,
Indices, Currencies, Softs and Meats. Tick sizes, margin rates and commissions
chosen according to Interactive Brokers LLC requirements and pricing structure.
Risk ratio was taken 2% per trade of total capital. Backward adjusted data series
were used for analysis. Backward adjusted data uses the actual prices of the most
recent contract with a backward correction of price discontinuities for successive
earlier active delivery months [8]. Considering that trading costs consist of com-
missions and also possible trade execution slippage, the size of a slippage was 3
ticks on every trade.

4. Experimental results
The strategy was back tested applying MatLab software of technical computing.
The experimental results of the strategy are presented in the following figure.

a) b)

Fig.2 Total daily returns using trading strategy

Fig. 2 (a) shows the dynamics of total returns of the strategy. Total returns had
outperformed the benchmark (Commodity Research Bureau- CRB Index) change
and was nearly 250% at the end of testing period. Fig.2. (a) also shows that despite
huge fluctuations in CRB Index (during 2008 market crash) the trading strategy
generated rather stable results. These results confirm the capability of the fractal
formation application in different market conditions. Fig.2(b) shows a comparison
of daily returns, where the biggest daily drawbacks are only up to 4%, while some
profitable days generate over 10% returns.
 5 Conclusions
In the paper experimental research of trading strategy based on truncated frac-
tal formation and applied in futures market was presented. The research was car-
ried out on daily historical data of the most active futures contracts in US markets.
The results have given significantly better returns if compared to CRB index. The
research has also shown rather stable results considering global economic fluctua-
tions over the testing period. The strategy can be attractive for hedge funds or fu-
tures market participants who intend trading short term strategies or implementing
risk management techniques, especially when volatility in markets increases.

Acknowledgment
This research as Fellowship is being funded by the European Union Structural
Funds project ”Postdoctoral Fellowship Implementation in Lithuania” within the
framework of the Measure for Enhancing Mobility of Scholars and Other Re-
searchers and the Promotion of Student Research (VP1-3.1-ŠMM-01) of the Pro-
gram of Human Resources Development Action Plan.

References
[1] Acworth Will (2011) “Record Volume 2010 (Annual Volume Survey)”, Futures Industry,
March 2011: 12-29.
[2] Avellaned M. (2002). Quantitative Analysis in Financial Markets. Collected Papers of the
New York University Mathematical Finance Seminar, Vol.3. World Scientific Pub Co Inc,
p.400
[3] Friesen, G.C., Weller P.A., Dunham L.M. (2009), Price trends and patterns in technical anal-
ysis: a theoretical and empirical examination. Journal of Banking&Finance, Vol. 33(6): 1089-
1100.
[4] Hautch N. (2008). Capturing common components in high-frequency time series: A multivar-
iate stochastic multiplicative error model. Journal of Economic Dynamics and Control,
Vol.32(12): 3978-4015.
[5] Izumi K., Toriumi F., Matsui H. (2009). Evaluation of automated- trading strategies using an
artificial market. Neurocomputing 72(16-18): 3469-3476.
[6] Mandelbrot B.B., Gomory R.E., Cootner P.H., Fama E.F., Morris W.S., Taylor H.M. (2010).
Fractals and Scaling in Finance: Discontinuity, Concentration, Risk. Springer, p.561.
[7] Masteika S. (2010) “Short term trading strategy based on chart pattern recognition and trend
trading in Nasdaq Biotechnology stock market”, Business Information Systems Workshops,
Springer, May 2010, pp.51-57
[8] Masteika, S., Rutkauskas A.V., Alexander Janes A. (2012). Continuous futures data series for
back testing and technical analysis. Conference proceedings, 3rd International Conference on
Financial Theory and Engineering. IACSIT Press, vol.29, Feb 2012, pp. 265-269.
[9] Miffre J., Rallis G. (2007). Momentum strategies in commodity futures markets. Journal of
Banking&Finance, Vol.31(6): 1863-1886.
[10] Szakmary A.C., Shen Q., Sharma S.C. (2010). Trend-following trading strategies in com-
modity futures: A re-examination. Journal of Banking & Finance, Vol.34(2): 409-426.
[11] Williams B.M., Williams J.G. (2004), Trading chaos: maximize profits with proven tech-
nical techniques, 2nd edition, John Wiley & Sons, Inc., p.228.

View publication stats