A robust portfolio optimization in Indian Stock market

Abstract

A good investment strategy requires a combination of mathematical modeling with deep understanding of the economics of the market. The basis of the portfolio optimization is the mean-variance optimization put forwarded by Markowitz in 1952. The optimization procedure depends on the input parameters, the covariance matrix and expected return which have to be estimated using the historical data. The portfolio selection hence depends on the reliability of these inputs and often lead to wrong results due to inaccurate estimation of covariance matrix and expected return. In this paper, we examine the performance of portfolio optimization in Indian Stock market using stable models for covariance estimation and come up with a portfolio of stocks that gives a meaningful return in reality.