A Study of Stock Dynamism in Asian Emerging Markets after the 2008 Economic Crisis Deng-Yiv Chiu* Department of Information Management, Chung Hua University, Hsinchu, Taiwan, ROC Email: chiuden@chu.edu.tw Agus Andria Department of Information Management, Chung Hua University, Hsinchu, Taiwan, ROC Email: d09803024@chu.edu.tw Abstract— The global financial crisis at the end of 2007 financially influenced various countries, including Indonesia. Because Indonesia achieved highest growth in the Southeast Asia region during the recession, global investors shift their investment in Indonesia market. Therefore, it is very important to explore the stock dynamism in Indonesia. We propose a hybrid approach of fuzzy theorem, support vector regression, genetic algorithm, and seasonal moving window to explore the Indonesian stock quarterly dynamism among the same quarter in continuous years using daily prices from 2006 to 2011. We find that the proposed method outperforms benchmark returns. We conclude that a hybrid approach is able to improve earning rate performances. Index Terms— emerging market, fuzzy c-means, genetic algorithm, moving window, support vector regression I. INTRODUCTION The global financial crisis at the end of 2007 triggered by the collapse of major financial institutions in the U.S. began to take effect in various countries. In Indonesia, the economic crisis in the U.S. has forced investors of institutional U.S. Treasury to release their holdings in the Indonesian capital market to strengthen the liquidity of financial institutions. The value of the shares was dropped and the volume of sales of shares in the capital market in Indonesia was reduced. Jakarta composite index (JCI) was down in the fourth quarter of 2007 and it continued until the end of 2008. It happened to global market also. Although economic growth slowed considerably during the recession, Indonesia achieved higher growth compared to the other G20 members with GDP 4.5% in 2009 [1]. Stable GDP growth is supported by domestic demand. Indonesia with 240 million of population is the fourth most populated country in the world which more than half of that population is under the age of 30. High growth GDP is followed by high investment in infrastructure. The Indonesian government has been eager to boost infrastructure financing up to 5% of GDP as well as attract private investment, which recorded US$ 941.5 million investment for infrastructure and US$ 711 million for mining sector in 2010. Considering global economy few years back, it appears that most of the foreign investors shift their investment from developed countries, like the U.S. to emerging countries in Asia, including Indonesia. Therefore, it is very important to explore the dynamism of the Indonesian stock market. Previous research has examined the relationship between intermediaries, stock markets, and real activity in four East Asian countries, including Indonesia [2]. A recent study shows that the trading rules have the stronger predictive power in the emerging stock markets than in the more developed stock market [3]. In the past decade, various methods have been widely applied to explore the internal dynamism of the stock market. Genetic algorithm (GA) is an approach used to avoid local optimum. GA simulates the revolution in biology to keep better chromosome to reach the purpose of optimization. Some research applies the optimized search property of the GA algorithm to locate distribution centers for single product network such that the sum of facility location, pipeline inventory, and safety stock costs is minimized [4]. Ref. [5] combines the vector autoregression (VAR) and genetic algorithm (GA) with a neural network (NN) to model and forecast Asian Pacific stock markets. Their results show that their system is more robust and makes more accurate predictions than the benchmark NN. Support vector machine (SVM) became a useful and popular method used by many researchers to avoid local optimum and achieve significant performance. Some research proposes a dynamic fuzzy model to explore the stock market dynamism. The fuzzy method combines various factors with an influential degree as the input variables, uses a GA algorithm to adjust the range of influential degree for each variable, and employs SVM to explore the stock market dynamism. The variables used in the experiment include technical indicators and macroeconomic variables [6]. Support vector regression (SVR) is extended from SVM. It adopts loss function and penalty parameter to avoid the effect of noise and outlier. SVR can convert nonlinear problems into high dimensional space and obtain good classification performance. Support vector regression is also used along with the fuzzy theorem in * Corresponding author: Deng-Yiv Chiu E-mail: chiuden@chu.edu.tw © 2013 ACADEMY PUBLISHER 2147 © 2013 ACADEMY PUBLISHER doi:10.4304/jsw.8.9.2147-2154