Application of Markov chain analysis to trend prediction of stock indices

The paper concerns with study aims at trying to predict the stock index trend of Prague stock exchange PX using Markov chain analysis (MCA). The prediction of the trend using MCA is done using time series of day closing prices from Jan.5, 2004, till Dec.12, 2009. Downloaded data are processed in order to provide current price rates and simple technical indicator, as well. Discrete state spaces are defined for four MCA models, and appropriate transition probability matrices are calculated. These objects represent a core of any MCA and its application. The results of the short-term trend prediction using MCA are reported for various investment strategies. Numerical calculations and computer implementations have been done by MS-Excel and Mathematica modules which are briefly discussed as well.