Working capital management is one of the strategies to achieve the optimal liquidity for the company. This study tests the effect of accounting variables and characteristics of peer firms (companies in the same industry) on working capital management strategy. The statistic sample consists of listed companies in Tehran Stock Exchange during the years 2009 to 2015. In order to test hypothesis, hierarchical Bayes model, Gibbs sampling and Monte Carlo simulations - Markov chain is used. The results show that the sector (industry) type and peer firms' stock returns average have a significant relationship with working capital management.

The multilevel models are used in applied sciences including social sciences, sociology, medicine, economic for analysing correlated data. There are various approaches to estimate the model parameters when the responses are normally distributed. To implement the Bayesian approach, a generalized version of the Markov Chain Monte Carlo algorithm, which has a simple structure and removes the correlations among the simulated samples for the fixed parameters and the errors in higher levels, is used in this article. Because the dimension of the covariance matrix for the new error vector is increased, based upon the Cholesky decomposition of the covariance matrix, two methods are proposed to speed the convergence of this approach. Then, the performances of these methods are evaluated in a simulation study and real life data.

In this work, the Markov Chain Monte Carlo is applied to estimate parameters that represent mechanisms that describe particles' dynamics in particulate systems from the literature's proposed models. Initially, the reduced sensitivity coefficient is evaluated to verify which parameters could be estimated simultaneously. The technique is then applied to estimate the models' parameters in different numerical scenarios to determine the rates that influence population dynamics. After the analyzes are performed, the estimates show good precision, accuracy, and a good fit between the measured and estimated state variables. The results show that the Markov chain Monte Carlo can determine the rates of population balance phenomenon.

This study, attempts to estimate and compare four different models of jumpdiffusion class combined with stochastic volatility that are based on stochastic differential equations, and their parameters latent variables are estimated by Markov chain Monte Carlo (MCMC) methods. In the Stochastic Volatility with Correlated Jumps (SVCJ) model, volatilities are scholastic, and the term jump is added to both scholastic prices and volatilities. The results of this study showed that this model is more efficient than the others are, as it provides a significantly better fit to the data, and therefore, corrects the shortcomings of the previous models and that it is closer to the actual market prices. Therefore, our estimating model under the Monte Carlo simulation allows an analysis on oil prices during certain times in the periods of tension and shock in the oil market.

There are some parameters in hydrologic models that cannot be measured directly. Estimation of hydrologic model parameters by various approaches and different optimization algorithms are generally error-prone, and therefore, uncertainty analysis is necessary. In this study we used DREAM-ZS, Differential Evolution Adaptive Metropolis, to investigate uncertainties of hydrologic model (HEC-HMS) parameters in Tamar watershed (1530 km2) in Golestan province. In order to assess the uncertainty of 24 parameters used in HMS, three flood events were used to calibrate and one flood event was used to validate the model. The results showed that the 95% total prediction uncertainty bounds bracketed most of the observed data especially peak discharge values but the uncertainty due to other sources than parameter uncertainty (e.g. forcing data (rainfall) and model structure error) are significant. Coefficient of variation for curve number (CN) was small for all flood events, therefore this parameters is more sensitive than the others. Histograms of the posterior probability density functions (pdfs) show that most of the individual parameters are well-defined and occupy only a relatively small region of the uniform prior distributions. Best simulation under DREAM-ZS was obviously better than simulation results of Nelder and Mead search algorithm.