Stock market index analysis with entropy approach

Analyzing and predicting the movement of the stock price index is one of the issues that analysts and investors face it and use various tools to do it. Considering the similarity of financial markets with physical phenomena, it is possible to study the relations existing in market as a complex system. One of the concepts in this area is entropy that measures the uncertainty and complexity of the dynamic system. In this research, the behavior of the Tehran Exchange Divedend and Price Index (TEDPIX) has been analyzed using Shannon's multiscale entropy technique. For this purpose, at the beginning, using the close price of stocks of Tehran Exchange's companies during the period from 2013 to 2017, entropy is calculated in monthly, seasonal, six-month, and annual periods, and in two scale, 50-50. Then, the existence of a grained causality relationship between these series and the TEDPIX was investigated using the Toda-Yamamoto test. The results of this study indicate that some Entropies are the linear cause of the stock index. In other words, the main information in the annual, six-month and seasonal periods, and small fluctuations in the seasonal period, is the linear cause of the TEDPIX.