Diffusion entropy analysis and random matrix analysis of the Indian stock market

Abstract

We analyze daily prices of 42 stocks listed in the Nifty50 index of National Stock Exchange of India from 2006 to 2019. The Nifty50 is one of the leading indexes of Indian the stock market which comprises 50 dominant companies. We use well established techniques such as random matrix theory, complex network analysis, hierarchical clustering, diffusion entropy, and conditional entropy to study the correlation among stocks and topological properties of the Indian stock market. We perform the static and dynamic analysis of the Nifty50 stocks to extract useful information from them. In the static period analysis, we consider three sub periods i.e. before, during and the subprime crisis of 2008. In dynamic period analysis, we consider overlapped and non-overlapped time windows of one year with varying shifts. We study the effect of ‘global events’ (such as the sub-prime crisis of 2008 and the Chinese crisis) and ‘local events’ (such as General elections 2009, Assembly elections in 7 states, General elections 2014 and Demonetization in India). We find that global and local extreme events affect the correlations among Nifty50 stocks of the Indian stock market. The analysis performed in this paper may be helpful in the development of different indicators of systemic risks in India.