Journal of Banking Financial Services and Insurance Research
Open Access
  • Year: 2011
  • Volume: 1
  • Issue: 1

An empirical study on stock index trend prediction using markov chain analysis

  • Author:
  • S. Vasanthi, M.V. Subha, S. Thirupparkadal Nambi
  • Total Page Count: 20
  • Page Number: 72 to 91

*Providence College for Women, Coonoor, The Nilgiris.

**Anna University of Technology, Coimbatore, Jyothipuram, Coimbatore 641 047.

***Guruvayurappan Institute of Management, Navakkarai, Palakkad Main Road, Coimbatore 641 105.

Abstract

The Indian Stock Market has been in action for quite some time and is in the limelight amidst others in the global capital markets. We can also see that the stock market indices around the globe are soaring to unprecedented heights and some markets are found to be encouraging investment destinations. Typically stock market indices are the performance indicators for the entire market. They act as barometers, which enable us to get an idea about the performance of the entire market in general. Interested investors all around the globe keep track of the movement of various stock market indices to get an idea about how the global markets are moving. With the globalization of the capital markets, we see that investors are not restricting their investments to a single market. Investments are made in different markets, enabling investors to enjoy the benefit of diversification of markets. Investors base their investments and make a choice of their investment destinations based on the movements of the stock markets. Investors worldwide try to watch the movement of market indices and have always shown keen interest in trying to predict the share market trend.

The present study aims at trying to predict the stock index trend of various global stock indices using Markov Chain Analysis. “A Markov Chain is a special kind of stochastic process where the outcome of an experiment depends only on the outcome of a previous experiment”. In other words, the next state of the system depends only on the current state and not on the previous states. Stochastic processes are of interest for describing the behavior of a system evolving over a period of time. In this study the First Order Markov Chain Model is applied to indices of various stock exchanges round the globe. Indices from markets like the American stock markets (DJIA, S&P 500), European Markets (FTSE, FTSH), Australian markets (AUSTA^ORD), China (SSE^), South East Asian markets (Hang Seng), Pakistan (KSE), India (BSE, NSE) etc. are chosen for the study. All major stock market indices representing popular investment destinations are included in the study. The results of the trend prediction using Markov Chain analysis is compared with the results obtained through traditional trend prediction tools. The prediction of the trend using Markov Chain Model is done using short term (one year data), medium term (3 year data) and long term (5 year data) and the results are compared. The results of the study show that majority of the time, Markov model outperforms the traditional trend prediction methods.