Design of Portfolio using Multivariate Analysis

Abstract

Stock markets are considered a barometer of the respective country#x2019;s economy around the world. Modern portfolio theory advocates diversification for risk management, which helps maintain returns as long as indices around the world are not perfectly correlated. The relationship exists across markets; as a result, co-movement has drawn the attention of individual investors and portfolio managers for the construction of their portfolios to maximize returns for a given level of risk. The study of co-movements provides inputs for portfolio construction and facilitates the identification of markets where indices may move in the same direction or the opposite direction and the country#x2019;s stock markets that are not correlated. A review of the literature revealed that statistical tools like Correlation, Factor analysis, and Granger causality test, etc., are some of the tools that can be used to understand co-movements of markets. Alan harper et al. (2012) study used principle component analysis and inferred that Indian stock returns are aligned with its trading partners and concluded that maximizing the investors#x2019; returns by reducing the risk. Tak Kee Hui concluded that factor analysis provides inputs for selecting foreign markets for risk diversification. This study examines the potential for diversification using 22 world stock market indices using multivariate analysis.