عنوان مقاله [English]
Bubbles are one of the most destructive factors of any market. In bubble literature, there are a variety of statistical and economic methods available to diagnose. These models generally suffer from math complications. In this paper, a simple statistical model is presented to identify speculative bubbles in stock markets. Since the parameters estimates of model are time varying, including transition probabilities, it is possible identify when and how newly born bubbles grow and burst over time. The parameters of the model can be estimated by recursive relations, however, since they require a huge storage capacity for computers, approximation in the computation are introduced which maintains the recursive nature of estimations. We then apply this model to the stock markets of the United States, Japan, and China and Iran. Advantages of this model are its simplicity, recursive relations, approximations and the use of Bayesian inference. Empirical results show the efficiency of this model in diagnosing the speculative bubbles.
JEL: G23, G32
How to cite this paper: Habibi, R., Salehi, M. R., & Zarepoor, M. (2017). Bayesian Modeling Speculative Bubbles in theStock Market in Iran. Quarterly Journal of Risk Modeling and Financial Engineering, 2(2), 225–241. (In Persian)