metropolis-hastings algorithm

Monte Carlo Markov chain methods are a set of algorithms for sampling possible distributions based on building a Markov chain with desirable properties. One of the most common Markov Monte Carlo chain algorithms is the Metropolis-Hastings algorithm. Therefore, the purpose of this study is to evaluate the relationship between financial distress and stock returns using the Metropolis-Hastings algorithm in the Tehran Stock Exchange. For this purpose, 151 companies were selected from the Tehran Stock Exchange in the period 2011 to 2020 using systematic elimination sampling method. R software was used to test the research hypotheses. Altman Z and Olson's score were also used to calculate financial distress. Also, in evaluating the relationship using the Metropolis-Hastings algorithm, two different previous distributions for the research variables were used. The results of the study showed that for Altman's financial distress variable, the accuracy of estimating financial distress was higher with the previous non-informed distribution. For O-Olson's financial helplessness variable, the precision of financial distress was higher with Zelner's previous distribution. Meanwhile, in the previous non-informed distribution, the effect of financial distress was not significant and was significant in Zelner's previous distribution.