حرکت مشترک شورای بازارهای سهام در حال ظهور همکاری خلیج فارس : شواهد جدید از آنالیز انسجام موجک
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|16019||2014||11 صفحه PDF||سفارش دهید||8740 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Economic Modelling, Volume 36, January 2014, Pages 421–431
This paper examines the short term and long term dependencies between stock market returns for the Gulf Cooperation Council (GCC) Countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates) during the period 2005–2010. Our empirical investigation is based on the wavelet squared coherence which allows us to assess the co-movement in both time-frequency spaces. Our results reveal frequent changes in the pattern of the co-movements especially after 2007 for all the selected GCC markets at relatively higher frequencies. We further note an increasing strength of dependence among the GCC stock markets during the last financial crisis signifying enhanced portfolio benefits for investors in the short term relative to the long term. On the financial side, we uncover that the strength of co-movement between GCC markets may impact the multi-country portfolio's value at risk (VaR) levels. These findings provide potential implications for portfolio managers operating in the GCC region who are invited to consider co-movement through both frequencies and time when designing their portfolios.
The behavior of stock market co-movement is a crucial issue in finance as it has important practical implications for portfolio's allocation and hedging strategy design. This issue has received much attention from academic researchers and practitioners since the works of Grubel, 1968 and Levy and Sarnat, 1970, and Longin and Solnik, 1995 and Longin and Solnik, 2001…, among others. It is well recognized that the main underlying idea of portfolio theory is connected to the advantages derived from sector and geographic diversification. Meanwhile, during the last two decades, international capital flows have been spectacularly increased subsequent to rapid mutations in global financial markets such as technological innovations and financial markets liberalization. All these factors have considerably raised the degree of stock market integration and promoted empirical research focused on international stock market co-movement. On the empirical side, this main research issue has been basically apprehended in the international finance literature and various empirical methodologies including cointegration approach (Arouri et al., 2011), error correction models, univariate and multivariate ARCH/GARCH-type models (Lin et al., 1994, Theodossiou and Lee, 1993, Chiang et al., 2007 and Ho and Tsui, 2003; Conrad et al., 2010, Aloui et al., 2011 and Sedik and Williams, 2011) rolling bi-correlation tests (Lim et al., 2008), and copula theory (Ye et al., 2012, Rodriguez, 2007, Aloui et al., 2011 and Samarakoon, 2011), were implemented to shed light on stock market co-movement and risk assessment. Overall, they concluded that stock market co-movement is not constant over time. Moreover, some scholars provided strong evidence of increasing international co-movement of stock returns (Brooks and Del Negro, 2006, Forbes and Rigobon, 2002, Karolyi and Stulz, 1996, Lee, 2004, Longin and Solnik, 1995 and Longin and Solnik, 2001). However, the distinction between short and long-term investor behavior should be considered in a co-movement analysis. Indeed, from a portfolio's diversification view, portfolio's managers are more interested in stock price co-movement at higher frequencies, that is, short-run movements. By contrast, other investors are concerned with lower frequencies (i.e. long-rum co-movement). Therefore, it would be very useful for portfolio's managers to resort to frequency domain in order to provide better understanding of stock market co-movement behavior at the frequency level. On the other hand, stock market co-movement has been intensively studied by means of various empirical methodologies but less much attention has been paid to the co-movement analysis in the frequency domain. In this vein, major previous works were devoted to developed stock markets (Madaleno and Pinho, 2010, Rua and Nunes, 2009 and Sharkasi et al., 2005; Vacha and Barunik, 2012, Garham and Kiviaho, 2012 and Garham and Nikkinen, 2011). Despite its recognized utility, analysis in the frequency domain is much less found in the financial and economic empirical literature. The first implementations of the wavelet approach are in economic in order to explain some linkages between several macroeconomic variables (Ramsey and Zhang, 1996 and Ramsey and Zhang, 1997). Also, Ramsey and Lampart (1998a) employed the wavelet to analyze the interactive relationship between several macroeconomic variables. Karuppiah and Los (2005) used this empirical approach to investigate the co-movement between some Asian exchange rates during the 1997 Asian crisis. Again, wavelet is employed by Nikkinen et al. (2011) to investigate the cross dynamic of exchange rate expectations. Concerning wavelet implementations in finance, main recent works applied the wavelet analysis for assessing the volatility transmission between the main developed stock markets. In fact, wavelets are considered as a powerful mathematical tool for signal processing which can provide more insights to co-movement among international stock markets via a decomposition of the time series into their time scale component. From a financial perspective, it is well documented that stock market co-movement can also lead to market contagion. Following Forbes and Rigobon (2002), financial contagion is defined “as a significant increase in cross-market linkages after a shock to an individual country”. More precisely, there's contagion only if markets show significant increase in co-movement during crisis period compared with periods of stability. If cross-market co-movement does not increase significantly after the shock, then any continued level of market correlation can be viewed only as interdependence between the two economies. In this study, we are concerned with GCC stock market co-movement. Our main research objective is to contribute to the literature of stock market co-movement using the Morlet wavelet coherence approach which allows us to analyze the frequency components of the stock market time series without losing the time information. Put it another way, the use of the wavelet approach allows us to detect stock market interactions which are hard to test out using any other modern econometric time-series models. The wavelet tools aim at studying the chronological specifications for financial and economic variables. Particularly, the decomposition into sub-time series and the localization of the interdependence between time series are the two most widely considered area of the wavelet approach in finance. As we know, markets consist of traders operating in different time horizons and therefore these traders can behave differently depending on dissimilar time resolutions (daily, monthly, and weekly). Thus, in terms of portfolio management, by analyzing the multi-scale dynamics of time series, the wavelet analysis appears useful for traders in order to analyze their investment horizons in different frequency bands of scale when they make management portfolio decisions. Furthermore, we should recognize that very less attention has been paid to emerging stock markets. Moreover, there's no empirical research founded on the wavelet coherence analysis and reserved to the stock market co-movement in the GCC region. Indeed, major previous studies employed standard time series econometric methods which consider the frequency and time components separately. As far as we know, this is the first empirical work implementing the continuous wavelet squared coherence to explore the dynamic linkage among the GCC stock markets in the frequency domain. The paper is structured as follows. Section 2 exposes the related literature and research motivations. Section 3 describes the wavelet coherence approach. Section 4 provides a detailed description of the data. Section 5 relates the results of the wavelet analysis results while Section 6 provides some managerial implications relative to the risk assessment for a GCC multi-country. The paper is closed with summary and some concluding comments.
نتیجه گیری انگلیسی
The assessment of the international stock market co-movement is considered as one of the major debated issue in empirical finance so as to shed light, on the potential benefits of international portfolio diversification, asset allocation and risk management. In this paper, we implemented the wavelet squared coherence approach to provide a fresh new look to the co-movement of the GCC stock markets. A noteworthy finding of this research can be summarized as follows: The wavelet coherence analysis revealed that co-movement depends on both frequency and time and is strongly affected by the occurrence of financial crisis. More importantly, the wavelet coherence analysis showed frequent changes in the pattern of the co-movement especially after 2005 and 2008 for all the selected GCC markets at relatively higher frequencies. For some countries such as Bahrain and Kuwait, the dynamic correlation between the two stock markets is high (more than 0.8) at high frequency (more than 217 days) which can be viewed as higher degree of persistence of shock transmission during turbulent periods. Qatar stock market seems to be unrelated to all the five markets for all investment horizons as well as the sample period. However, stock markets in Saudi Arabia and U.A.E. exhibit strong co-movement mainly during the sharp price dives. These findings conjectured that turmoil periods create fear in the GCC markets which sequentially increases the downside risk for portfolios founded on GCC listed firms. The wavelet squared coherence approach also uncovered the change in co-movement to relatively higher frequency overlaps with the inception of the recent financial crisis. For instance, the occurrence of the subprime financial crisis has to a large extent increased the degree of co-movement between all the GCC stock markets. From a financial perspective, the increasing of GCC stock market coherence during the historical financial crisis periods especially at high frequencies corroborates the “contagion hypothesis” during these periods. From an empirical outlook, the time-varying behavior of the correlation coefficients could result in structural beaks in the asset price series when significant external shocks occurred. Our findings visibly indicate the changing pattern of co-movement among the GCC markets and they may offer several implications for portfolio managers, international investor as well as for the hedge funds operating in the GCC region. Conclusively, our findings pave the way for several new research extensions. First of all, it would be interesting to implement the wavelet squared coherence analysis to other stock markets. Secondly, our results may exhibit some structural breaks which may induce that some GCC stock markets may perhaps be independent of certain crisis or even benefit from crises elsewhere. This point of view may be argued by the flow of capital from the crisis market to the non-crisis market and further empirical investigations in this issue are surely preeminently needed. Finally, it could be useful for portfolio managers operating in the region to assess the delay in response stock market volatility shocks indicating possible occurrence of arbitrage opportunities.