The paper considers the Bayesian analysis of the threshold stochastic volatility models. Studies of methods for analyzing stochastic volatility and improving models of stochastic volatility significantly improve the quality of forecast models and their estimates. Bayesian inference is performed by tailoring Markov chain Monte Carlo (MCMC) or sequential Monte Carlo (SMC) schemes that take into account the specific characteristics of models. The results of applying the method demonstrated in models heteroscedastic non-stationary processes.