印度市場與全球市場之聯結

全文

(1)⊕ 國立中山大學企業管理學系研究所碩士論文. 印度市場與全球市場之聯結 The Interlinkage of India market with World Market. 研究生：尤吉塔撰指導教授：陳月霞博士. 中華民國九十八年一月.

(2) Department of Business Administration, College of Management National Sun Yat-Sen University Master Thesis. 印度市場與全球市場之聯結 The Interlinkage of India market with World Market. Yogita Jayendra Kadam Student ID: M954011054 Advisor: Prof. Dr. Yueh H. Chen 2008-09. This Thesis is submitted in part fulfillment of the requirements for the degree of Master of Business Administration. .

(3) . . ACKNOWLEDGEMENTS. I would like to express my gratitude to Dr. Yueh H. Chen for her invaluable supervision during the development of this thesis. I would like to thanks Dr. Liao and Dr. Chen, the two members of my thesis oral test committee, to endorse my thesis. I would also like to express my thanks to the Department of Business Administration, College of Management, National Sun Yat-Sen University for providing me with the necessary background through the MSBA program. Special thanks go to my husband Jayendra for his endless support and help throughout my school years and my son Shivam for his support by less troubling me while working on thesis.. Yogita Jayendra Kadam National Sun Yat-Sen University, MSBA 2008-09. 2 .

(4) . . The Interlinkage of India Market with World Market Yogita Jayendra Kadam Advisor: Prof. Dr. Yueh H. Chen. Abstract The relationship between the stock markets of the developed countries has been examined extensively in the literature. This paper examine the dynamic relationship between India and the major developed markets including USA, UK and Japan .Using daily stock market data from January 1997 to December 2002 and from January 2003 to December 2007,the study examine the stock price indices of India (BSE SENSEX), USA (Dow Jones Industrial Average), UK(FTSE-100) and Japan (Nikkei 225). The ordinary least square method is showing some relationship between the stock markets. A multiple equation series known as a vector autoregression is proposed for describing the dynamic behavior of the four stock markets. The result shows that the markets are interrelated at significant level and influences each other. All the markets influence India but recently the influence of USA market is comparatively high than other developed markets. Keywords: Stock prices, Vector Autoregression.. 3 .

(5) . . 印度市場與全球市場之聯結指導教授陳月霞教授 Yogita Jayendra kadam. 中文摘要過去有關國家間的股票市場的關係已有廣泛探討，本文主要檢驗印度與主要已開發國家，如美國、英國與日本的動態關係。本文使用指數分別為印度BSE SENSEX指數、美國Dow Jones工業指數與英國FTSE-100指數與日本Nikkei-225，樣本期間為1997年1月至2002年12月與2003年1月至2007年12月的每日資料。本文OLS實證顯示在各市場間具有共整現象，且多重線性迴歸使用向量自我回歸顯示在這四個市場裡具有動態行為且彼此影響著，但最近尤以美國市場較其他市場具有影響性。. 關鍵詞:股票價格，向量自我回歸. 4 .

(6) 印度市場與全球市場之聯結. 指導教授陳月霞教授 Yogita Jayendra kadam. 中文摘要. 過去有關國家間的股票市場的關係已有廣泛探討，本文主要檢驗印度與主要已開發國家，如美國、英國與日本的動態關係。本文使用指數分別為印度BSE SENSEX指數、美國Dow Jones工業指數與英國FTSE-100指數與日本Nikkei-225，樣本期間為1997年 1月至2002年12月與2003年1月至2007年12月的每日資料。本文OLS實證顯示在各市場間具有共整現象，且多重線性迴歸使用向量自我回歸顯示在這四個市場裡具有動態行為且彼此影響著，但最近尤以美國市場較其他市場具有影響性。. 關鍵詞:股票價格，向量自我回歸. 4.

(7) . . TABLE OF CONTENTS 1. Introduction 2. Literature Review 3. Types of Stock Price Indices 3.1 BSE-SENSEX (India) 3.2 Nikkei-225(Japan) 3.3 Dow Jones Industrial Average (USA) 3.4 Financial Times Stock exchange - 100 (UK). 4. Data Analysis and Research Methodology 4.1 Data Analysis 4.2 Research Methodology 4.2.1 Unit Root Test 4.2.2 Ordinary Least Square Method 4.2.3 Vector Autoregression (VAR) and Identification. 5. Empirical Results Table 5.1 Descriptive Statistics of Stock Price Indices for the Period Table 5.2 Correlation Matrix of Stock Return between Stock Exchanges Table 5.3 Unit Root Analysis with log level and first difference Table 5.4 Ordinary least square test results of stock market indices Table 5.5 presents the result of Lag order selection criteria Table 5.6 VAR (Vector Autoregression) Test Results. 6. Research Limitations/Implications 7. Conclusions and Suggestions. 5 .

(8) . . 1. Introduction It is well known that the movements in the stock prices are influenced by the flow of market information. One possible source of this information is movement in other stock markets in the world. There are many reasons why the returns of two or more different markets might be related. The markets belonging to different economies might be related through trade and investment, so that any news about economies fundamentals of one country permeates to another and thus affect each other’s equity markets. Financial integration expresses the links between financial markets. It can be classified as total, direct and indirect integration. Direct financial integration, also referred to as capital market integration, is expressed in deviations from ‘the law of one price’ for financial securities. In other words, when there is direct integration the investor can expect the same risk-adjusted return on investments on different markets. Indirect financial integration refers to a situation in which the return on an investment in one country is indirectly linked to the return on investments in other countries. A stock market return represents the equity market. The co-movements in the indices can be used to estimate integration. The globalization of capital flows has led to the growing relevance of emerging capital markets; India is one of the countries with an expanding capital market that is increasingly attracting funds from the foreign countries. Actually, in line with the global trend, reform of the Indian stock market began with the establishment of Securities and Exchange Board of India (SEBI) in 1988 to frame rules and guidelines for various operations of the stock exchange in India. Nevertheless, the reform process gained momentum only in the aftermath of the external payments crisis of 1991 followed by the securities scam of 1992. Among the significant measures of opening up capital market, portfolio investment by foreign indirect investors (FIIs) such as pension funds, mutual funds, investments trusts, asset management 6 .

(9) . . companies, nominee companies and incorporated portfolio managers allowed since September 1992 have made the turning point for the Indian stock markets. As of now FIIs are allowed to invest in all categories of securities traded in the primary and secondary segments and in the derivatives segment. Attracting foreign capital appears to be the main reason for opening up of the stock markets for FIIs. Progressively the liberal policies have led to increasing inflow of foreign investment in India, both in terms of direct investment and portfolio investment. In general, the deregulation and market liberalization measures and the increasing activities of multinational companies will continually accelerate the growth of Indian stock market. Given the newfound interest in the Indian stock markets, an intriguing question is how far India has gone down the road towards international financial integration, and whether the linkages exist among the stock indices of India and world’s major stock indices. To find the answer of these questions, the research examine the interrelationship between Indian stock markets and major developed stock markets and study the underlying mechanism through which the Indian stock indices interact with international stock indices by analyzing empirically the long-run the pair wise Unit Root test, OLS and VAR test between the Indian stock market and the world major developed markets including US, UK and Japan. The remainder of the paper is organized as follows: Section 2 provides a overview of empirical literature on the inter-linkages and interaction of equity markets, Section 3 describes types of stock price indices used in this study, Section 4 data analysis and research methodology adopted in the study, Section 5 discusses the empirical results, Section 6 describes the limitations/Implication of the study and Section 7 provides conclusions and suggestions. (Source: India, Ministry of Finance, Economic Survey) 7 .

(10) . . 2. Literature Review The globalization of capital flows has led to the growing relevance of emerging a capital market. Early works in this field came up with strong evidence of inter linkage between the stock markets around the globe, a result of global economic integration. The interest in the interdependency between the global stock markets heightened after the global market crash of October 1987. Only a few studies have examined the co-movement of Indian stock market with international markets. Sharma and Kennedy (1977) conducted a test of the price behavior of Indian market with the US and UK markets and found that the behavior of the Indian market is statistically impossible to differentiate from that of the US and UK markets and found no evidence of systematic cyclical component or periodicity for these. Taylor and Tonks (1989) study the bi-variate cointegration technique (by Engle & Granger, 1987) with a view to make research regarding the market integration concerning markets of US, Germany, Netherlands and Japan. They used monthly data on stock price indices for the two sub periods: April 1973 to September 1979, and October 1979 to June 1986.The result shows a) former period: No cointegration between the stock price return of these countries. b) Later period: cointegration between the stock price return of the UK with the stock price return of the US, Germany, Netherlands and Japan. They suggested the absence of long-term gain from diversification for the UK investors after the abolition of exchange controls. Rao & Naik (1990) applied Cross-Spectral analysis to examine the inter-relatedness of US, Japanese and Indian Stock Markets. The monthly data of stock price indices used in the study were of monthly type and those were collected from New York, Tokyo and Bombay stock 8 .

(11) . . exchanges for the period January, 1971 to December, 1988. They tried to cover the episode of the October, 1987 world market crash and the Indian stock market boom of 1985-87 and subsequent fall in 1987-88 in their study period. They found that for the Indian stock return, the gains estimates are ‘independent’ from the US or Japan indices. They reflect the institutional fact that the Indian economy has been characterized by heavy controls throughout the entire seventies with liberalization measures initiated only in the late eighties and concluded that the interrelationship among the three markets were, in average, very low. Cheung and Ng. (1992) cited the dynamic properties of stock returns in Tokyo and New York, using daily close-to-close market indices from January 1985 to December 1989. GARCH types of model were used to describe the inter-temporal behavior of these stock indices. They included foreign market return’s lagged return in the mean equation and lagged square return in the variance equation of the home market model to capture mean and volatility spillovers. They find that the US market is an important global factor from January 1985 to December 1989. Lee and Kim (1994) ) cited evidence for a significant increase in the co movement of the stock price indices after the crash. The national stock market became more interrelated and the co-movement among national stock markets was stronger when the US stock market is more volatile. By adopting Bayesian methods Koop (1994) analyzed unit root and cointegration properties of two different finance data sets concluded that there are no common trends in stock prices across countries. Choudhury (1994) study the relationship among the Asian Newly Industrialized Economies NIEs, Japan and the US. He found that the US led the NIEs and that there were significant linkages between the markets using variance decomposition and impulse response functions. 9 .

(12) . . Lin, Engle & Ito (1994) examined how returns and volatilities of stock indices are correlated between the Tokyo, and New York stock markets. They selected the opening price return, 30 minutes (in New York) or 15 minutes (in Tokyo) after the market officially opens, covering a period of 4 years, starting from October 1985 to December 1989.They used intra-daily data to decompose daily returns into daytime and overnight returns and analyzed international transmission mechanism by applying two models, one was the Aggregate Stock Model and the other was Single-extraction Model. After estimating the above models, they made them compared with the GARCH-in-Mean model of Hamao, Masulis and Ng. (1990). They found that (a) the foreign daytime returns can significantly influence the domestic overnight returns. Contrary to that belief that the New York stock returns influence Tokyo, but not the vice versa, they found that cross market interdependence was generally bi- directional between New York and Tokyo markets; (b) there was little evidence of the lagged return spillovers from New York day time to Tokyo day time and vice versa. Kee-Hang and Andrew Karalyi (1994) examine the joint dynamics of overnight and day time return volatility for the Nikkei Stock Average in Tokyo and the Standard & Poor’s 500 Stock Return in New York over a period of 4 years starting from 1988 to 1992. They proved that the magnitude and persistence of shocks originating in New York or Tokyo that transmit to the other markets were significantly understated while ignoring the asymmetric effects. Arshanapalli, Doukas and Lang (1995) examine the possible links and dynamic interactions between the US and six major Asian Stock Markets before and after October 1987. By considering daily closing stock market return time series, adjusted to account for the time zone differences during the period for January 1986 to May 1992, they empirical results proved presence of a long run equilibrium relationship between the US and Asian stock market movements during the post October 1987 period. Their co integration results, based on the Asian equity markets alone, supported the possibility of increased regional capital market 10 .

(13) . . integration among the six Asian stock exchanges during the post crash period. However, their error-correction analysis, at the regional level, failed to support the presence of a strong cointegrating relationship among the Asian markets. Lastly, they concluded that the Asian equity markets were less integrated with Japanese equity market than they were with the US market. Corhay, Rad and J. Urbain (1995), study the stock markets of Australia, Japan, Hong Kong, New Zealand and Singapore and find no evidence of a single stochastic trend for these countries. Choudhury, (1997) conducted empirical investigation of the long-run relationship between stock indices from six Latin American markets and the United States, by means of unit root test, Johansen method of co-integration tests and error correction models using the log of weekly data from January, 1989 to December, 1993 .The unit root test result showed the evidence of a stochastic trend in all indices. The co-integration tests showed the presence of a long-run relationship between the six Latin American indices (with and without the US return) and the error correction results proved the significant casualty among the stated indices. Francis and Leachman (1998) combine the Johansen (1988) procedure for cointegration testing with tests of weak exogeneity and invariance in order to ascertain whether a system of equity markets is characterized by superexogeneity. Their studies reveal that US stock market influences other markets around the world. The world’s second largest economy is of Japan but still it does not influence other markets. Janakiramanan and Lamba (1998) examine the linkages between the stock markets in the Pacific-Basin region using daily market indices during the period 1988-96. They used VAR model to trace the relationships between developed markets of Australia, Hong-Kong, Japan, New Zealand, Singapore and the US, as well as from developing markets of Indonesia, 11 .

(14) . . Malaysia and Thailand, those markets. Their examination result showed that the US market influenced all other Australian markets, except Indonesia, and none of these markets exert a significant influence on the US market. By excluding US market for the VAR system, they found persistent linkages between the markets considered. They also found significant interrelationship among markets those are geographically and economically close and/or have large numbers of cross-border listings. Lastly, they concluded that the markets closing earlier in the day exerting greater influence over markets closing later in the day. Liu, Pan and Shieh (1998) examined the stability of the interrelationship among the emerging and developed stock markets of Thailand, Taiwan, Japan, Singapore, Hong-Kong and the US. They divided the samples into two sub-samples: January 2, 1985 to October 16, 1987 and October 19, 1987 to December 31, 1990. They found that after the 1987 crisis there is an increase in the general stock market interdependence and interdependence within the Asian Pacific regional markets. Masih & Masih (1999) studied the long- and short-term dynamic linkages among international and Asian emerging stock markets, and also conducted a test to quantify the extent of the Asian stock market fluctuations, by intra-regional contagion effect. The eight stock market indices, such as, Thailand, Malaysia, the US, the UK, Japan, Hong-Kong, and Singapore stock market indices considered, using end-of-day national stock price indices from 14 February 1992 to 19 June 1997. They applied some recent time-series econometric techniques, such as Vector Error-Correction Model (Toda and Phillips, 1993), level VAR model including integrated and co integrated processes of arbitrary orders (Toda and Yamamoto, 1995). They conclude that US is the leader at the global level for short as well as long term, and there is a significant relationship between the OECD and the Asian emerging markets.. 12 .

(15) . . Kumar (2002), by using weekly closing data sets of stock indices from 1994 to 1999 attempted to find out the attractiveness of the Indian stock market from the investors in the developed markets like US, Japan, Singapore and Hong-Kong. The Johansen’s Maximum Likelihood Method is used to show the interrelationship between the stock market indices of the said countries. The result showed that there was no long run relationship between those markets. stock return of Indian stock market was not cointegrated with that of developed markets and the Indian stock market does not fell any disturbance occurred in major developed markets. Mishra (2002) tried to investigate the international integration of India’s domestic financial markets by studying the international integration of Indian stock market. Stock indices from Bombay Stock Exchange as well as from NASDAQ Stock Exchange were taken into account from the period 1993-94 to 1999-2000. By applying Ordinary Least Square and Cointegration technique, they found a positive correlation between NASDAQ and BSE. They conclude that BSE was being influenced by the movements of NASDAQ. But there is no co integrating vector between BSE and NASDAQ indices which shows that there was no longrun relationship between these two stock exchanges. Pretorius (2002) took data for analysis from 10 emerging stock markets, namely Argentina, Brazil, China, Greece, India, Korea, Malaysia, Mexico, South Africa and Turkey, from the first quarter of 1995 to the third quarter of 2000. They made their analysis in two parts, one is cross-sectional analysis, and the other is time-series approach. In the first, all the pair-wise simple correlations were polled across all the country pairs and then regressed on the averages of the possible explanatory variables in order to explain the causes as well as the degree of the stock market correlationship. In their second part, they followed a time-series approach to explain why such relationships changes over time. Their results showed that only the extent of bilateral trade and the industrial production growth differential were significant in explaining 13 .

(16) . . the correlation between two countries on a cross-sectional basis. While, the results of the time-series regression showed that only the extent of bilateral trade, IP growth differentials, a dummy to reflect the 1998 emerging market crisis and regional dummy variables were significant to explain the pair-wise correlation coefficients Kiran Kumar & Mukhopadyay (2002) examine the short run dynamic linkages between NSE Nifty in India and NASDAQ composite in the US, during the period 1999 to 2001. By using intra-daily data to determine the daytime and overnight returns, a comprehensive study is carried out to examine the co movement and volatility transmission between the US and Indian stock markets. Their study employed a two-stage GARCH model along with an ARMA-GARCH model to specify the impact of the NASDAQ composite daytime return and volatility on not only the mean but also the conditional volatility of Nifty overnight returns. The simple ARMA-GARCH model is used to show a better performance. They have a series of important findings: (a) The Granger causality results supported a unidirectional granger causality running from the US stock market to Indian stock market; (b) The NSE Nifty overnight return of a day is significantly affected by the previous day’s day time return of both NASDAQ composite and NSE Nifty. They conclude that the volatility spillover effects were significant only from NASDAQ composite. Nath and Verma (2003) studied the transmission of market movements among three major stock markets in Asian region, viz., India, Singapore and Taiwan, during the period 1994 to 2002. They used daily data relating to stock market indices of India (NSE NIFTY), Singapore (STI) and Taiwan (TAIEX). The bi-variate and multivariate cointegration analysis (Granger 1969,1988 and Johnson 1988)test result shows that there was no long-term interrelationships and thus an equilibrium among those stock markets, though they confirmed the possibilities, in few cases, of some casual influences of one stock market’ return on the return in other. 14 .

(17) . . stock markets. The international investors could achieve long term gains by investing in the stock markets because of the independencies of the stock markets. Danny I. Cho and Tomson Ogwang (2004) conducted test for cointegration between the TSX Composite Return and the TSX Venture Composite Return of Canadian stock prices using two different unit root tests using daily data covering the period from December 10, 2001 to February 6, 2004.. The results shows that the two series are not cointegrated indicating that there is no significant long run relationship between them, although each series has a unit root, however, there is evidence of unidirectional causality from the TSX Composite Return to the TSX Venture Composite Return. Varsha Kulkarni and Nivedita Deo (2005) conducts a test of volatility of an Indian stock market in terms of aspects like participation, synchronization of stocks and quantification of volatility using the random matrix approach. Volatility pattern of the market is found using the BSE return for the three-year period 2000-2002, using daily returns of 70 stocks for several time windows of 85 days in 2001 to do a brief comparative analysis with statistics of eigenvalues and eigenvectors of the matrix C of correlations between price fluctuations, in time regimes of different volatilities. The test result shows that the deviations from RMT bounds are more pronounced in volatile time periods. The largest eigenvalue, which is in some sense an return of true information in the entire market, is seen to be highly sensitive to the trends of market activity. Hafiz Al Asad Bin Hoque (2007) used Cointegration model to examine the long-run equilibrium relationship among the time series. There is evidence of cointegration among the variables. The results demonstrate that stock prices in Bangladesh, USA, Japan and India share a common stochastic trend. He detected long-term relationship and short-term dynamics in his study, which confirms financial liberalization in Bangladesh since 1990, has 15 .

(18) . . successfully opened up Bangladesh stock market towards the outside world and hence the stock market is influenced by other markets. Numerous number of studies has tested the relationships among international stock markets . In the current changing economic scenario of India , a comprehensive study to explore the inter linkage between India and equity market of developed world is required to explore the opportunity of investment diversification.. 3. Types of Stock Price Indices 3.1 BSE-SENSEX (India): The Bombay Stock Exchange Sensitive Return or BSE 30 is a "market capitalisation-weighted" return of 30 stocks representing a sample of large, wellestablished and financially sound companies. The SENSEX is calculated using a market capitalisation-weighted methodology. As per this methodology, the level of return at any point of time reflects the total market value of 30 component stocks relative to a base period. 3.2 Nikkei-225 (Japan): Nikkei -225 is a stock market return for the Tokyo Stock Exchange (TSE). The Nikkei average is the most watched return of Asian stocks. It is a price-weighted average (the unit is Yen), and the components are reviewed once a year. 3.3 DJIA (USA): Dow Jones Industrial Average is a price-weighted return based on 30 stocks. The companies incorporated in the calculations of the return right now are major factors in their respective industries or sectors, and their stocks are widely held by individuals and institutional investors. A stock is typically added only if it has an excellent reputation, demonstrates sustained growth, is of interest to a large number of investors and accurately represents the sector(s) covered by the average. 3.4 FTSE-100 (UK): The Financial Times Stock Exchange -100 is the most widely quoted and popular return for tracking the London stock exchange. The return comprises of the 16 .

(19) . . shares of the top 100 U.K. companies ranked by market capitalization. FTSE-100 is a market capitalization-weighted return, re-weighted every day.. 4. Data Analysis and Research Methodology 4.1 Data Analysis Basically, a stock market price return is a portfolio of individual stocks. The return level corresponds to some average of the price levels of individual shares. Changes in the return level give rise to market returns. Thus, the stock market return can commonly be use as an indicator of the market performance. A stock market return can be viewed simply as a portfolio of shares. The stock market return of a country may also be an indicator of shortterm portfolio investment movement in the country. An upward trend of a stock market return means that there is an increase in demand of the listed shares in the market. This indicated that investors are attracted to buy shares and invest their fund in the country. On the other hand, a downward trend movement of a stock market return indicates that the investors are unlikely to continuously hold the listed shares. Therefore, stock market movements may reflect the attractiveness of a country for investments, especially for portfolio investments. In this research, the daily indices of the stock exchanges of the four stock market, which are BSE SENSEX (India), Dow Jones Industrial Average (United States), FTSE 100 (United Kingdom) and Nikkei225 (Japan) are used to measure the countries daily stock market movement .The data covers a total period of 11 years starting from 1st January 1997 through 31st December 2007 consisting daily stock indices. The India time is 5.5 hours ahead of Greewichmean time (GMT) and there is no overlapping of stock market time. For better results the data are divided into two parts: first one starts from 1997 and ends at 2002 and second part starts from 2003 and ends on 2007. The indices are adjusted to be in terms of US. 17 .

(20) . . dollars for better comparison. The continuously compounded rate of return is calculated by using the following formula Rt = ln (Pt /Pt-1) Where Rt = Return on day ‘t’ Pt = Index value on day ‘t’ Pt-1 = Index value on day ‘t-1’ ln = Natural log. 4.2 Research Methodology There are several methods for testing the flow of information and co-movement of prices in stock markets across the countries. In this study the emphasis is given to test the inter-market relationship among the stock market in India with that of equity markets of developed world, via; (i). Descriptive statistics (ii) Correlation matrix (iii) Unit root test (iv) Ordinary Least Square Method and (v) VAR test.. 4.2.1 Unit Root Test Two time series are cointergrated when a linear combination of the time series is stationary, even though each series may individually be a non-stationary. Since non-stationary time series do not return to their long run average values following a disturbance, it is important to convert them to stationary processes otherwise regressing one non-stationary process can generate spurious results. If a time series contains a stochastic trend, it is said to be integrated of order d, i.e. I(d) and differencing the series d times yields a stationary series. Since the market series are likely to be I(1) processes they will tend to be non-stationary in levels but stationary in first differences.. 18 .

(21) . . The method used to analyze the daily market returns involves the vector autoregression (VAR) model proposed by Sims(1980) which allows to analyze the transmission of market movement across countries. The method requires the time series to be stationary. The stationary of time series has been examined by using unit root tests. Augmented Dickey Fuller (ADF–1979, 1981), Phillips-Perron (1987, 1988) has been employed for the said purpose.. The Augmented Dickey Fuller test examines the presence of unit root in an autoregressive model. A simple AR (1) model is yt = ρyt − 1 + ut. (1). Where, yt is the variable of interest, t is the time return, ρ is a coefficient, and ut is the error/disturbance term. The regression model can be written as Δyt = (ρ−1) yt−1+ ut = δyt − 1 + ut. (2). Where, Δ is the first difference operator. This model can be estimated and testing for a unit root is equivalent to testing. δ = 0. The Dickey-Fuller tests assume that the error terms are statistically independent and have a constant variance which is rather strict assumption so an alternative test, the Phillip-Perron test, is also used to test the stationarity of data. It may be noted that Phillip-Perron test allows the error disturbances to be weakly dependent and heterogeneously distributed. yt = α0 + α1 yt-1 + αt {t- T/2} + ut. (3) 19 . .

(22) . . Test statistics for the regression coefficients under the null hypothesis that the data are generated by yt = yt-1 + ut, where E (ut) = 0. A financial time series is said to be integrated of one order i.e. I (1), if it becomes stationary at the first difference. If two series are integrated of order one, there may have a linear combination that may be stationary without differencing. If said condition fulfills then these are called cointegrated.. 4.2.2 Ordinary Least Square Method In this method multiple regression model with k = 3 independent variables is adopted. The equation is as follows: INt = α0+α1JPt+α2UKt+α3USt+ut. (4). Where α0 is the intercept, α1, α2 and α3 are the slope parameters of returns of the independent variables. Ut is the error term or disturbance.IN= INDIA (SENSEX), JP= JAPAN (NIKKEI225), UK = UK (FTSE- 100) and US= USA (DJIA) variables to be examined. Conditional expectation: E (u/ α1, α2, α3) =0. (5). The (5) equation requires that all factors in the unobserved error term be uncorrelated with the explanatory variables.. 4.2.3 Vector Autoregression (VAR) and Identification The trend of each macroeconomic variable was examined before VAR estimation procedures were done. Using a time series plot of each variable, the movement of a particular variable return was described as it fluctuated significantly at various times. In its simplest form, a 20 .

(23) . . vector autoregression is an unrestricted reduced form model that expresses each variable as a linear function of a constant and the lags of that and each other variable in the system.. Vector autoregressive (VAR) model is a systems regression model, which contains more than one dependent variable. In VAR all variables are endogenous and symmetrically treated. The nonstructural VAR model is estimated on a set of (endogenously determined) variables. In this case the endogenous variables are stock price returns of India, Japan, UK and USA. The four equations of VAR (m) are given below, the m represents number of lags (vectors). INt = α1 + α11INt-1 + α12JPt-1 + α13UKt-1+ α14USt-1 + α11INt-2 + α12JPt-2+ α13UKt-2. + α14USt-2 +v1t. JPt = α2 + α21INt-1 + α22JPt-1 + α23UKt-1+ α24USt-1 + α21INt-2 + α22JPt-2+ α23UKt-2 + α24USt-2 +v2t UKt = α3 + α31INt-1 + α32JPt-1 + α33UKt-1+ α34USt-1 + α31INt-2 + α32JPt-2+ α33UKt-2 + α34USt-2 +v3t USt = α4+ α41INt-1 + α42JPt-1 + α43UKt-1+ α44USt-1 + α41INt-2 + α42JPt-2+ α43UKt-2 + α44USt-2 +v4t. (6). Where, IN= INDIA (SENSEX), JP= JAPAN (NIKKEI-225), UK = UK (FTSE- 100 ) and US= USA (DJIA ) returns of variables to be examined. αi (i=1,2,3,4) is constant term or intercept vector, α11through α44 are n×n coefficient matrices, and vit (i =1,2,3,4) is a the corresponding n×1 disturbance vector.. 21 .

(24) . . In matrix notation four-equation VAR (m) model is written compactly as: Yt = α0 +β (L) Yt+vt. (7). Here we let Yt be 4×1 vector of stock price returns (IN, JP, UK AND US respectively).L simply denotes a polynomial in the lag operator. α0 is the constant term vector, and vt represents the corresponding disturbance vector, thus the right hand side of the equation (6) contains only past values of four stock price returns as well as constant and error terms.. 5. Empirical Results Table 5. 1: Descriptive Statistics of Returns of Stock Price Indices for the Period Panel A: 1997-2002 JAPAN 126.4839 124.3411 202.6478 67.91362 32.94036 0.056156 1.976314. UK 8626.506 8668.581 11209.60 5695.758 1290.319 -0.191941 1.894872. USA 9450.616 9720.760 11722.98 6391.690 1305.782 -0.425270 2.038621. Jarque-Bera 87.19005* 69.15646* Probability 0.000000 0.000000 Observations 1565 1565 * Indicates significant at 1% level.. 89.24886* 0.000000 1565. 107.4417* 0.000000 1565. JAPAN 115.8790 111.8766 153.6145 63.30517 24.45139 -0.301465 2.025348. UK 9606.274 9331.861 13870.63 5282.602 2184.438 0.137911 2.018748. USA 10889.98 10599.99 14164.53 7524.060 1455.988 0.281252 2.706699. Jarque-Bera 184.1672* 71.36521* Probability 0.000000 0.000000 Observations 1304 1304 * Indicates significant at 1% level.. 56.44872* 0.000000 1304. 21.86568* 0.000018 1304. Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis. INDIA 87.82797 86.90283 136.1159 54.26450 18.41661 0.311823 2.026262. Panel B: 2003-2007 Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis. INDIA 201.1692 165.1457 517.9546 61.90172 111.1537 0.918116 3.133544. 22 .

(25) . . Descriptive statistics for the stock price returns are given in Panel A and B of Table 5.1. The result shows that the mean, standard deviation of all the stock indices are positive. The level of volatility is measured by standard deviation. All the stock indices are skewed to the right except the UK, USA (1997-2002) and Japan (2003-2007) stock indices which skewed to the left. The kurtosis all the indices are less than 3 except for India (2003-2007). Under the null hypothesis of a normal distribution, the Jarque-Bera statistic is distributed as with 2 degrees of freedom. The reported Probability is the probability that a Jarque-Bera statistic exceeds (in absolute value) the observed value under the null hypothesis. A result of probability value leads to the rejection of the null hypothesis of a normal distribution at the 1% significance level.. Table 5.2: Correlation Matrix of Stock Return between Stock Exchanges Panel A.1997-2007 1997-2002 IN INDIA (IN). JP. UK. US. 1.000. JAPAN (JP). 0.899*. 1.000. UK. (UK). 0.511*. 0.612*. 1.000. USA. (US). 0.075*. 0.125*. 0.557*. 1.000. Panel B. 2003-2007 2003-2007 IN. JP. UK. INDIA (IN). 1.000. JAPAN (JP). 0.862*. 1.000. UK. (UK). 0.949*. 0.941*. 1.000. USA. (US). 0.941*. 0.900*. 0.970*. US. 1.000. * Indicates significant at 1% level. Panel A and B of Table 5.2 presents the pair wise correlation coefficient for the stock indices .The correlation coefficients are in the range of .075 to .899. (1997-2002) and .862 to .970 (2003-2007). From the table it is clear that for the period 1997-2002 India is highly correlated with Japan; whereas correlation between India and USA is low at 1% significant level. For the period 2003-2007, India is highly correlated with all stock indices at 1% significant level. The results of previous studies imply that increase in correlation among the 23 .

(26) . . world equity market reduces the diversification benefits. The correlation between Japan and USA is very low. The low correlation between markets shows that foreign investors can achieve substantial risk diversification benefits. Correlation analysis is weak technique as it does not discuss the cause and effect relationship. In order to take a better picture of the affairs, the OLS and VAR tests are performed to tests the flow of information and comovement of prices in stock markets across the countries after testing the stationarity of stock price indices. Table 5.3 Unit Root Analysis 5.3 A. The log levels of stock market indices Constant Country 1997-2002 INDIA. ADF -1.60(0). Constant and Trend PP. -1.69 (7). ADF -2.21(0). PP -2.32(6). JAPAN -1.07 (0) -0.98 (4) -1.56(0) -1.47(4) UK -1.42 (0) -1.07 (23) -2.07(0) -1.77(24) USA -2.42 (0) -2.36 (9) -1.91(0) -1.78(9) 2003-2007 INDIA 1.87(1) 2.20(33) -0.54(1) -0.29(32) JAPAN -1.70(0) -1.66(10) -2.72(0) -2.57(5) UK -0.65(1) -0.62(20) -4.10(1)* -4.17(9)* USA -0.92(1) -1.00(24) -2.75(1) -2.78(20) 5.3 B. The log levels at first difference of stock market indices.. Country 1997-2002 INDIA JAPAN UK USA 2003-2007 INDIA JAPAN UK USA. Constant Δ ADF Δ PP. Constant and Trend Δ ADF Δ PP. -37.51(0)*. -37.50 (9)*. -37.51(0)*. -37.49(9)*. -41.39(0)* -26.18(2)* -39.40(0)*. -41.43(4)* -40.57(25)* -39.48(10)*. -41.37(0)* -26.27(2)* -39.44(0)*. -41.42(4)* -41.06(27)* -39.58(12)*. -31.05(0)* -36.86(0)* -40.05(0)* -39.30(0)*. -30.91(38)* -37.09(10)* -40.80(18)* -39.75(23)*. -31.17(0)* -36.86(0)* -40.03(0)* -39.28(0)*. -30.84(42)* -37.13(11)* -40.79(18)* -39.73(23)*. 24 .

(27) . . Note: 1.DF is the Dickey –Fuller t-statistic; ADF is the augmented Dickey-Fuller statistics and PP is Phillips –Perron test statistic. 2. *, **, *** indicates significance at the 1%, 5% and 10% respectively. 3. The numbers within brackets shows for DF and ADF statistics indicates Lag length based on SIC. 4. The numbers within brackets shows for PP statistics represents the bandwidth selected based on Newey-West method using Bartlett Kernel. 5. Δ indicates the log First Difference. Table 5.3A, the constant and constant and linear trend is included in the test equation. The results show that all the four the stock price indices exhibit units root by not rejecting the null hypothesis except the stock of UK. The results show that for UK stock indices (2003-2007), the ADF and PP test are showing the stationarity at 1% significant level at constant and trend level by rejecting the null hypothesis. Thus all the four stock indices behave as random walks except that of UK (2003-2007) under ADF and PP test. The unit root test results for Table 5.3B, the logs of the first differences of the series rejected the null hypothesis that all the four series are non stationary under ADF and PP tests at 1% significant level. Table 5.3 A and B results shows that the hypothesis of non-stationarity in the market indices cannot be rejected. If residual for a I(1) is stationary the linear combination of it could say that the stock prices are integrated. Table 5.4 Ordinary least square test results of stock market returns A. For the period from 1997-2007. Variable C JP UK US R-squared. Dependent Variable: IN 1997-2002 Coefficient Std. Error t-Statistic 29.9320 1.6308 18.3542 0.5224 0.0082 63.4627 -0.0007 0.0002 -3.1457 -0.0001 0.0001 -0.7341. Prob. 0.0000 0.0000 0.0017 0.4630 0.811323 25 . .

(28) . . Adjusted R-squared Durbin-Watson stat F-statistic Prob(F-statistic). 0.810960 0.054447 2237.463 0.000000. B. For the period from 2003-2007 Dependent Variable: IN 2003-2007 Variable Coefficient Std. Error t-Statistic Prob. C -347.8072 13.2483 -26.2528 0.0000 JP -1.1013 0.1103 -9.9793 0.0000 UK 0.0459 0.0022 20.4504 0.0000 US 0.0216 0.0026 8.2529 0.0000 R-squared 0.914672 Adjusted R-squared 0.914475 Durbin-Watson stat 0.031155 F-statistic 4645.108 Prob(F-statistic) 0.000000. The ordinary least square result shows that the estimated coefficients are statistically significant. The coefficient of the C is the constant or intercept in the regression—it is the base level of the prediction when all of the other independent variables are zero. The slope quantifies the steepness of the line. It equals the change in Y for each unit change in X. It is expressed in the units of the Y-axis divided by the units of the X-axis. If the slope is positive, Y increases as X increases. If the slope is negative, Y decreases as X increases. For 19972002, the slope is showing 29.93209 positive results and for the period 2003-2007 it is showing a negative result of -347.8072. For the period 1997-2002 Japan is showing positive result and for the period 2003-2007 it is showing negative result. Whereas UK and USA are showing negative results for the period 1997-2002 and positive result for 2003-2007, which state that it will adversely affect India returns. When there is increase in the prices of UK and USA returnss, the price of India return will be decrease. The value of R2 is 0.811323(199726 .

(29) . . 2002) and 0.914672(2003-2007) i.e. it is close to 1 which means the regression model is perfectly fit and there is strong relationship between all variables. The Durbin-Watson statistic shows that there is evidence of positive serial correlation in the residuals .As the pvalue is less than the significance level 0.05, we reject the null hypothesis that all slope coefficients are equal to zero. This method is not reliable as sometimes the results may not provide the accurate information. In this study based on the findings USA, UK and Japan are interrelated and they are not independent. To get more accurate and better result VAR test is performed. VAR Lag Order Selection Criteria: A vector autoregression was estimated where the four equations corresponded to stock price returns of India, Japan, UK and USA. A constant term was included in each equation. A two lag specification is chosen in the VAR since this lags length minimizes the Akaike information criterion (AIC). Harvey (1994) provides more details on fitting or specifying the VAR model as well as the selection process used. Table 5.5 presents the result of Lag order selection criteria VAR Lag Order Selection Criteria (1997-2002) Lag. LogL. LR. FPE. AIC. SC. HQ. 0. -32283.24. NA. 4.07e+16. 49.596. 49.612. 49.602. 1. -20888.13. 22702.69. 1.04e+09. 32.116. 32.196. 32.146. 2. -20690.41. 392.7209*. 7.89e+08*. 31.837*. 31.980*. 31.891*. (2003-2007) 0. -39231.22. NA. 7.48e+16. 50.205. 50.218. 50.210. 1. -25452.29. 27469.71. 1.68e+09. 32.594. 32.662. 32.619. 2. -25311.66. 279.6355*. 1.43e+09*. 32.434*. 32.557*. 32.480*. 27 .

(30) . . * indicates lag order selected by the criterion. LR: sequential modified LR test statistic (each test at 5% level), FPE: Final prediction error, AIC: Akaike information criterion, SC: Schwarz information criterion and HQ: Hannan-Quinn information criterion.. The VAR test results are provided in Table 6 A and B, reveals the insight into the interdepencies of stock movement measured at four stock price indices. The t statistics shows significant results. A two lag specification is chosen in the VAR since this lag length minimizes the Akaike Information Criterion (AIC). For the period 1997-2002 the t statistic shows that India is affected by its own lags at 1% and 5% level. Japan and UK are showing influence on India at both the lags at 1% significant level for the period 1997-2002. For the period 2003-2007, the result shows that India is highly influence by USA markets at 1% significant level. Japan is affected by its own lags and also influenced by UK at both the lags for the period 1997-2002 and at lag one for the period 2003-2007 at 1% significant level. The USA is showing its influence on Japan from 1997-2007 at 1% significant level on both lags. The t statistics of UK shows that it is affecting its own lags and is influence by USA at 1% critical level on both the lags. The USA market is also affected by Japan at 10% level for lag one and UK at 1% and 10% level for the period 1997-2002. The result shows that India market does not have any impact on USA, UK and Japan market. There is evidence of strong interrelationship between the variables. The India market is less volatile and interdependent on USA market. The result shows that US market has significant influence on all the markets compare to other markets.. 28 .

(31) . . Table 5.6 VAR (Vector Autoregression) Test Results A. 1997-2002 IN 1.035047 (0.02517) [ 41.1159]*. JP 0.038140 (0.03330) [ 1.14537]. UK -2.980391 (1.69617) [-1.75713]. US 1.269581 (1.87280) [ 0.67791]. IN(-2). -0.059795 (0.02508) [-2.38377]*. -0.026933 (0.03318) [-0.81171]. 2.540087 (1.69013) [ 1.50289]. -1.983772 (1.86613) [-1.06304]. JP(-1). -0.059199 (0.01879) [-3.15060]*. 0.902623 (0.02485) [ 36.3163]*. -1.582777 (1.26602) [-1.25020]. -2.376926 (1.39786) [-1.70041]. JP(-2). 0.072638 (0.01880) [ 3.86340]*. 0.087122 (0.02487) [ 3.50310]*. 1.899510 (1.26681) [ 1.49944]. 2.702319 (1.39873) [ 1.93198]***. UK(-1). 0.001792 (0.00040) [ 4.52069]*. 0.002874 (0.00052) [ 5.47943])*. 0.916033 (0.02671) [ 34.2924]*. 0.071261 (0.02949) [ 2.41610]*. UK(-2). -0.001810 (0.00040) [-4.55806]*. -0.002761 (0.00053) [-5.25680]*. 0.080996 (0.02675) [ 3.02765]*. -0.065565 (0.02954) [-2.21967]***. US(-1). 0.000591 (0.00036) [ 1.62576]. 0.003867 (0.00048) [ 8.04446]*. 0.237426 (0.02449) [ 9.69532]*. 0.973676 (0.02704) [ 36.0103]*. US(-2). -0.000584 (0.00036) [-1.61013]. -0.003890 (0.00048) [-8.10946]*. -0.241373 (0.02443) [-9.87892]*. 0.017567 (0.02698) [ 0.65118]. `C. 0.542936 (0.35548) [ 1.52734]. -0.516831 (0.47021) [-1.09914]. 60.70081 (23.9514) [ 2.53433]*. 56.33471 (26.4456) [ 2.13021]**. R-squared Adj. Rsquared F-statistic Akaike AIC Schwarz SC. 0.992721 0.992684. 0.996012 0.995991. 0.993255 0.993220. 0.991932 0.991891. 26493.76 3.753689. 48508.35 4.313141. 28603.53 12.17433. 23883.26 12.37246. 3.784520. 4.343972. 12.20517. 12.40329. IN(-1). 29 .

(32) . . B. 2003-2007 IN 1.106512 (0.02920) [ 37.8994]*. JP 0.018334 (0.01178) [ 1.55593]. UK 0.958733 (0.82300) [ 1.16493]. US -0.800129 (0.74321) [-1.07659]. IN(-2). -0.108991 (0.02926) [-3.72488]*. -0.020046 (0.01181) [-1.69750]. -0.959208 (0.82481) [-1.16295]. 0.801070 (0.74484) [ 1.07549]. JP(-1). 0.001674 (0.06758) [ 0.02476]. 0.891107 (0.02728) [ 32.6694]*. -0.633832 (1.90509) [-0.33270]. -0.234699 (1.72039) [-0.13642]. JP(-2). -0.019560 (0.06758) [-0.28945]. 0.098933 (0.02727) [ 3.62741]*. 1.045956 (1.90490) [ 0.54909]. 0.028351 (1.72022) [ 0.01648]. UK(-1). -2.23E-05 (0.00114) [-0.01955]. 0.002574 (0.00046) [ 5.59486]*. 0.771308 (0.03214) [ 24.0020]*. 0.009737 (0.02902) [ 0.33552]. UK(-2). 0.000101 (0.00114) [ 0.08821]. -0.002449 (0.00046) [-5.31437]*. 0.215821 (0.03219) [ 6.70473]*. 0.000488 (0.02907) [ 0.01680]. US(-1). 0.010895 (0.00122) [ 8.95200]*. 0.005055 (0.00049) [ 10.2916]*. 0.284200 (0.03431) [ 8.28438]*. 0.912580 (0.03098) [ 29.4575]*. US(-2). -0.010391 (0.00122) [-8.51908]*. -0.004990 (0.00049) [-10.1362]*. -0.272312 (0.03438) [-7.92038]*. 0.074264 (0.03105) [ 2.39192]*. `C. -3.406964 (1.68717) [-2.01933]**. -0.398194 (0.68094) [-0.58477]. -48.78775 (47.5593) [-1.02583]. 72.90573 (42.9484) [ 1.69752]. R-squared Adj. Rsquared F-statistic Akaike AIC Schwarz SC. 0.999118 0.999113. 0.997023 0.997005. 0.998184 0.998173. 0.996663 0.996642. 183141.0 5.238554. 54138.22 3.423879. 88827.57 11.91640. 48271.21 11.71244. 5.274303. 3.459628. 11.95215. 11.74819. IN(-1). 30 .

(33) . . Note: The country codes of the stock price returns used are: IN- INDIA, JP - JAPAN, UKUK and US- USA. There are 2864 observations after adjusting endpoint to account for two lags. Standard errors are in ( ) & t-statistics are in [ ]. *.**, *** indicates significant at the 1%, 5% and 10% respectively.. 6. Research Limitations/Implications: The investor can have some diversification benefit even though the markets are correlated. The further understanding of market behavior can be enhanced by the investor through comparison with other studies with technical, fundamental analysis and also with financial anomalies (Fong et al 2005 and Wong et al 2005).. 7. Conclusions and Suggestions: The present study endeavored to explore dynamics of the stock price movement in the context of a developing country India. This study analyzed stock price movement and interrelationship between the India stock market and world developed stock markets by using the daily data of BSE SENSEX (India), DJIA(USA), FTSE-100(UK) and Nikkei-225(Japan) divided into two parts from 1997 to 2002 and 2003-2007 for better results. The unit root test shows that all the series are integrated at order I(1) showing stochastic trend. The correlation matrix result shows positive correlation between India and other stock indices. The Indian stock market is statistically significantly cointegrated with the stock markets in USA, UK and Japan by using OLS estimation. The VAR (m) model is used to examine the long run equilibrium relationship among the time series. There is evidence of interrelationship among the variables. The result shows the influence of USA market on India is more compare to other stock markets in recent years. The evidence of dynamic relationship helps investors in making efficient investment decisions in the India and other stock market. 31 .

(34) . . Appendix: Graphs showing the trend in the stock price indices 1. Figure shows India stock return for 1997-2002 and 2003-2007. 140 600. 130 120. 500. 110. 400. 100 300. 90 80. 200. 70. 100. 60 0. 50 1997. 1998. 1999. 2000. 2001. 2003. 2002. 2004. 2005. 2006. 2007. INDIA. INDIA. 2. Figure shows Japan stock return for 1997-2002 and 2003-2007. 220. 160. 200 140. 180 160. 120. 140 100. 120 100. 80. 80 60. 60 1997. 1998. 1999. 2000 JAPAN. 2001. 2002. 2003. 2004. 2005. 2006. 2007. JAPAN. 32 .

(35) . . 3. Figure shows UK stock return for 1997-2002 and 2003-2007. 14000 12000. 13000. 11000. 12000. 10000. 11000 10000. 9000. 9000 8000. 8000 7000. 7000. 6000. 6000 5000. 5000 1997. 1998. 1999. 2000. 2001. 2003. 2002. 2004. 2005. 2006. 2007. UK. UK. 4. Figure shows USA stock return for 1997-2002 and 2003-2007. 15000. 12000. 14000 11000. 13000 10000. 12000. 9000. 11000 10000. 8000. 9000 7000. 8000 7000. 6000 1997. 1998. 1999. 2000 USA. 2001. 2002. 2003. 2004. 2005. 2006. 2007. USA. 33 .

(36) . . References . 1. Arshanapalli, B., Doukas, J., Lang, L.H.P.( 1995), pre and post October 1987 Stock Market Linkages between US and Asian Markets, Pacific Basin Finance Journal 3, 57-73. 2. Arun K. Mishra (2002), International Financial Integration of Domestic Financial Markets: A Study of India, The ICFAI Journal of Applied Finance 8(2), 5-15. 3. Cha, B. and Oh, S. (2000), ‘‘The relationship between developed equity markets and the Pacific Basin’s emerging equity markets’’, International Review of Economics and Finance, Vol. 9, pp. 299-322. 4. Cheung & Lition K. Ng.( 1992), Interactions between the US and Japan Stock Markets Indices, Journal of International Financial Markets, Institutions & Money 2(2), 51-70. 5. Cheung, Yin-Wong and Lillian K. Ng (1992), “Stock price dynamics and firm size: an empirical investigation”, The Journal of Finance, Vol. 47, 1985-1997. 6. Choudhury, A.R.( 1994), Stock Market Interdependencies: Evidence from the Asian NIEs, Journal of Macroeconomics 16, 629-651. 7. Choudhury, T. (1997), Stochastic Trends in Stock Prices: Evidence from Latin American Markets, Journal of Macroeconomics 19, 285-304. 8. Corhay, A., A. Rad and J. Urbain (1995), “Long-run behavior of Pacific-Basin stock prices”, Applied Financial Economics, Vol. 5, pp.11-18. 9. Danny I. Cho, and Tomson (2004), Testing for cointegration and causality between TSX composite return and TSX venture composite return, Faculty of Business Brock University, 500 Glenridge Avenue, St.Catharines, Ontario, L2S, 3A1, June 2004. 10. Dickey, D. A. and W.A. Fuller (1979), “Distribution of the Estimators for Autoregressive Time Series with a Unit Root,” Journal of the American Statistical Association, 74, 427-432. 11. Dickey, D. A. and W.A. Fuller (1981), “Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root,” Econometrica, 49, 1057-1072. 34 .

(37) . . 12. Fong , W.M., H. Lean and W.K. Wong(2005), “Stochastic dominance and the rationality of the momentum effect across markets”. Journal of Financial markets, Vol. 8 No.1 p89-109. 13. Francis, B. and Leachman, L. (1998), ‘‘Superexogeneity and dynamic linkages among international equity markets’’, Journal of International Money and Finance, Vol. 17, p.475-92. 14. Hafiz Al Asad Bin Hoque (2007), Co-movement of Bangladesh stock market with other markets Cointegration and error correction approach. Emerald Group Publishing Limited, Vol. 33 No. 10, 2007pp. 810-820. 15. Hamao, Y., Masulis, R., and Ng, V. (1990), Correlations in Price Changes and Volatility across International Stock Markets, Review of Financial Studies 3, 281-307. 16. http://www.bseindia.com 17. Janakiraman, S. and A.S.Lamba( 1998), An Empirical Examination of Linkages Between Pacific-Basin Stock Markets, Journal of International Financial Markets, Institutions and Money 8, 155-173. 18. Kee-Hong Bae, Karolyi (1994), Good News, Bad News and International Spillovers of Stock Return Volatility between Japan and the US, Pacific-Basin Finance Journal 2, 405-438. 19. Kiran Kumar & Mukhopadya, (2002), A Case of US and India, Research Paper, NSEIndia. 20. Koop, Gary. (1994), “An objective Bayesian analysis of common stochastic trends in international stock prices and exchange rates”, Journal of Empirical Finance, Vol. 1, pp.343364. 21. Lee, S.B., and K.J. Kim (1994), “Does the October 1987 crash strengthen the co-movement in stock price returnes”, Quarterly Review of Economics and Business, Vol.3, No.1-2, p. 89102. 22. Lin, Engle & Ito (1994), Do Bull and Bears Move Across Borders? International Transmission of Stock Returns and Volatility, Review of Financial Studies 7, 507-538.. 35 .

(38) . . 23. Liu, Y.A., Pan, M.S., Shieh, J.C.P.( 1998), International Transmission of Stock Price Movements: Evidence from the US and five Asian-Pacific Markets, Journal of Financial Economics 22, 59-69. 24. Masih, A.M., Masih, R. (1999), Are Asian Stock Markets Fluctuation Due Mainly to Intraregional Contagion Effects? Evidence Based on Asian Emerging Stock Markets, PacificBasin Finance Journal 7, 251-282. 25. Nath, G.C. & Verma, S. (2003), Study of Common Stochastic Trend and Cointegration in the Emerging Markets: A Case Study of India, Singapore & Taiwan, Research Paper, NSEIndia. 26. Phillips, P.C.B. and P. Perron (1988), “Testing For a Unit Root in Time Series Regression,” Biometrika, 75, 335-346. 27. Pretarius, E. (2002), Economic Determinants of Emerging Stock Market Interdependence, Emerging Markets Review (March), 84-105. 28. Rao, B.S.R. & Umesh Naik, (1990), Inter-Relatedness of Stock Markets: Spectral Investigation of US, Japanese and Indian Markets, 309-321 29. Rao, B.S.R. and Umesh Naik (1990), “Inter-relatedness of stock market spectral investigation of USA, Japan and Indian markets note”, Artha Vignana, Vol. 32, p. 309-321. 30. S.S.S. Kumar, (2002), Indian Stock Markets in International Diversification: A FIIs Perspective. 31. Sharma, J.L. and R.E. Kennedy (1977), “Comparative analysis of stock price behavior on the Bombay, London & New York Stock Exchanges”, The Journal of Financial and Quantitative Analysis, Vol. 12, No. 3. (Sep. 1977), pp. 391-413. 32. Sheng, H.C. and Tu, A.H. (2000), ‘‘A study of cointegration and variance decomposition among national equity indices before and during the period of the Asian financial crisis’’, Journal of Multinational Financial Management, Vol. 10, p. 345-65.. 36 .

(39) . . 33. Taylor, M.P. & Tonks, I. (1989), The Internationalization of Stock Markets and Abolition of UK Exchange Control, The Review of Economics & Statistics 71, p332-336. 34. Varsha Kulkarni1 and Nivedita Deo (2005), Volatility of an Indian stock market: A random matrix approach, University of Delhi, Delhi 110007, India.19 Dec 2005.. 37 .

(40)