Today there are different methods for evaluation of financial markets, but yet predicting stock returns accurately is not so simple. From the past to the present, Auto Regressive models introduced by Box and Jenkins are used in predicting various issues like financial matters and the performance of these models shows that there are links between future values of the series and current and past values . Beside the modeling, the stock price on the stock exchange is always an interesting topic. In this context, modeling and forecasting stock returns using autoregressive process with two regimes are very useful.
This paper is seek to model the rate of return on financial indicators in relation with the transition probabilities in two distinct regimes. these models are also very useful for short term forecasting. Methods in This study is autoregressive Markov model (MS-AR).
It is mentioned that financial stock cycles can be described better with some models with two regimes than a simple AR model. MSAR is a non-linear model that can model jumps and the financial crisis well due to the breakdown of the data series.
Markov model can identify accurate years of regime change in Iran and explain the dates of shocks in detail. This model have the ability to manage separation of stock returns with low and high volatility regimes. The results show that the market remained in the high volatility regime for a longer period than low volatility regime. This is important because investors can make profit from the volatility of the profit opportunities identified