Fad Models with Markov Switching Hetroskedasticity: Decomposing Tehran Stock Exchange Return into Permanent and Transitory Components

In this paper, the stochastic behavior of Tehran stock exchange return index (TEDPIX) is examined by using unobserved component Markov switching model (UC-MS) during the period 3/27/2010 - 8/3/2015. In this model, stock returns are decomposed into two components; permanent and transitory components. The transitory component has three-state Markov switching heteroskedasticity (low, medium, and high). Results show that UC-MS model is appropriate for this data. Low value of RCM criteria implies that model can successfully distinguish between regimes from behavior of data. The sum of the autoregressive coefficients in temporary component indicates that 40 percent of current value of temporary component is explained by its 2-period lagged values. The duration of high-variance regimes for transitory component are short-lived and revert to normal levels quickly. The presidential election has significant effect on being in the third regime.