The Number of MCMC Draws Needed to Compute Bayesian Credible Bounds
نویسندگان
چکیده
Bayesian credible bounds produced from Markov chain Monte Carlo (MCMC) procedures contain Monte Carlo error and thus may require a long chain in order to have a reasonable degree of repeatability. This is especially true when there is a substantial amount of autocorrelation in the chain realization. Repeatability would be important in some applications where it would be undesirable to report numerical values containing substantial Monte Carlo error in the least significant digits. The endpoints of a credible interval correspond to quantiles of the empirical distribution of the MCMC draws from the marginal posterior distribution of the quantity of interest. Our goal is to provide an algorithm to choose the number of MCMC draws that will provide, with high probability, a specified amount of precision (i.e., the number of correct significant digits) in the interval endpoints.
منابع مشابه
Retrospective sampling in MCMC with an application to COM-Poisson regression
The normalisation constant in the distribution of a discrete random variable may not be available in closed form; in such cases the calculation of the likelihood can be computationally expensive. Approximations of the likelihood or approximate Bayesian computation (ABC) methods can be used; but the resulting MCMC algorithm may not sample from the target of interest. In certain situations one ca...
متن کاملBayesian Estimation of Change Point in Phase One Risk Adjusted Control Charts
Use of risk adjusted control charts for monitoring patients’ surgical outcomes is now popular.These charts are developed based on considering the patient’s pre-operation risks. Change point detection is a crucial problem in statistical process control (SPC).It helpsthe managers toanalyzeroot causes of out-of-control conditions more effectively. Since the control chart signals do not necessarily...
متن کاملA simulation approach to convergence rates for Markov chain Monte Carlo algorithms
Markov chain Monte Carlo (MCMC) methods, including the Gibbs sampler and the Metropolis-Hastings algorithm, are very commonly used in Bayesian statistics for sampling from complicated, high-dimensional posterior distributions. A continuing source of uncertainty is how long such a sampler must be run in order to converge approximately to its target stationary distribution. Rosenthal (1995b) pres...
متن کاملMarkov Chain Monte Carlo and Gibbs Sampling
A major limitation towards more widespread implementation of Bayesian approaches is that obtaining the posterior distribution often requires the integration of high-dimensional functions. This can be computationally very difficult, but several approaches short of direct integration have been proposed (reviewed by Smith 1991, Evans and Swartz 1995, Tanner 1996). We focus here on Markov Chain Mon...
متن کاملMarkov Chain Monte Carlo and Gibbs Sampling
A major limitation towards more widespread implementation of Bayesian approaches is that obtaining the posterior distribution often requires the integration of high-dimensional functions. This can be computationally very difficult, but several approaches short of direct integration have been proposed (reviewed by Smith 1991, Evans and Swartz 1995, Tanner 1996). We focus here on Markov Chain Mon...
متن کامل