Global Journal of Management and Business Research, A: Administration and Management, Volume 21 Issue 12

countries are Hong Kong, Indonesia, South Korea, Malaysia, Philippines, Singapore, Taiwan, and Thailand. In contrast, the European countries are France, Germany, Italy, the Netherlands, Portugal, Spain, Switzerland, and the United Kingdom. This gives us two groups of equal size for comparison purposes. Because we use groups of neighboring countries and the time zone difference within either group does not exceed 2, issues of lead-lag structure for studying volatility spillovers and covariance are not critical for us, and current data can be used for this study. The data covers five years from 12/31/93 to 12/31/98. Thus, we have 1305 data points in our sample. We consider 7/2/97 to be the beginning of the Asian crisis. In our study, we look at the correlation structure of the eight Asian countries and compare it with the correlation structure among the European countries. Salim M Darbar and Partha Deb (1997) investigated the co-movements of equity returns for indices of four major equity markets, namely Toronto 300 share index, Topix, the financial times stock exchange 100 shares, and S&P 500 index for a period of 1 Jan 1989 to 31 Dec 1992 by using Multivariate GARCH framework. The authors found that the US and Japan have transitory correlations, but there is no evidence of permanent correlation, and conditional correlations can change considerably in reaction to the news. Portfolios can be adjusted based on the variations shown in correlations. Michel Beine and Bertrand Candelon (2007) examined the impact of trade and financial liberalization on the degree of stock market co- movement among emerging economies using a sample of 25 developing countries observed over 15 years. The authors estimate the impact of reforms that aim at opening these countries to trade and financial channels to the rest of the world. Estimating time-varying cross- country correlations allows the econometric investigation to be performed using a panel data framework, raising the quality of the statistical inference. Our results offer strong support in favour of a positive impact of trade and financial liberalization reforms on the degree of cross-country stock market linkages. Ritesh Patel (2017) found that various investors like Foreign institutional investors, High net worth individuals, institutional investors, retail investors derive an advantage in diversifying the fourteen stock markets that the author considered in the study. The author used the Johnsen cointegration test to find the relationship among the selected stock markets and found a long- term relationship among the selected stock markets. The Granger causality test proved that BSE returns are affected by BVSP, FTSE-100, and MXX. The author suggested that investors take their portfolio investment decision by observing the long- and short-term markets integration Indian market. III. R ationale for the S tudy The literature review revealed that many studies had been carried out to understand the co-movements and integration of the various stock markets. The studies focused more on American, European, Asia, and Asia Specific markets, but limited work has been done covering Asia Pacific, Europe, American, and Middle East markets. Hence, the present study included middle east markets apart from the other world markets. The study's findings can be used by the retail and institutional investors, especially the Indian investors, for designing their portfolio and those who are seeking other than their markets for risk minimization. This study has been carried out to investigate the potential for diversification into various stock indices by using the concept of cointegration and co-movement among world stock markets. One of the well-known multivariate analysis techniques, like factor analysis, has been used in the study to understand the relationship among the latent variables. It obtains a reduced set of uncorrelated latent variables using a set of linear combinations of the original variables to maximize the variance of these components, and few studies have been conducted using the Principle Component analysis and Maximum Likely hood methods (Abbas Valadkhani et al. 2008). Factor analysis is used to determine the co-movements among the 22 selected world markets covering major continents like Asia, Asia Pacific, Europe, North America, America, Belgium, etc. Alan Harper and Zhenhu Jin's (2012) approach has been adopted for the present study. IV. M ethodology of the S tudy Monthly indices data for the period from1 Jan 2000 to 31 Dec 2018 are used in the present study. Twenty-two stock markets indices were considered for the study. They are ASX 200 (Australia), Nikkei 225 (Japan), KOSPI (South Korea), Hang Seng (Hong Kong), Jakarta Composite Index (Indonesia), SSE Composite Index (China), Taiwan Capitalization Weighted Stock Index, Sensex (India), Amman SE General (Jordan), BLOM(Lebanon), QE General (Qatar), MSM 30(Oman), Tadawul (Saudi Arabia), Tel Aviv (Israel), CAC 40 (France), DAX 30 (Germany), BEL 20 (Belgium), Euronext 100, DJIA (United States), TSE (Canada), BOVESPA (Brazil) and BMV (Mexico). The monthly returns have been calculated and tested their stationary by using ADF and PP tests. Factor analysis conceptually helps identify the variables that have similar patterns associated with the latent factor or variable. Factor analysis traditionally assumes that there is no unit root in time series data. This model is well known multivariate analysis model (Hair et al., 1998; Tabachnick and Fidell, 2001; Tsay, 2002). Similarly, the same concept can be used to identify the markets which are associated and not associated. This information is 16 Global Journal of Management and Business Research Volume XXI Issue XII Version I Year 2021 ( ) A © 2021 Global Journals Design of Portfolio using Multivariate Analysis