Exact Sampling from a Continuous State Space
نویسندگان
چکیده
Propp & Wilson (1996) described a protocol, called coupling from the past, for exact sampling from a target distribution using a coupled Markov chain Monte Carlo algorithm. In this paper we extend coupling from the past to various MCMC samplers on a continuous state space; rather than following the monotone sampling device of Propp & Wilson, our approach uses methods related to gamma-coupling and rejection sampling to simulate the chain, and direct accounting of sample paths.
منابع مشابه
Exact sampling for Bayesian inference: towards general purpose algorithms
Propp and Wilson (1996) described a protocol, called coupling from the past, for exact sampling from a target distribution using a coupled Markov chain Monte Carlo algorithm. In this paper we discuss the implementation of coupling from the past for samplers on a continuous state space; our ultimate objective is Bayesian MCMC with guaranteed convergence. We make some progress towards this object...
متن کاملExact Transition Probabilities for the Independence Metropolis Sampler
A recent result of Jun Liu's has shown how to compute explicitly the eigen-values and eigenvectors for the Markov chain derived from a special case of the Hastings sampling algorithm, known as the indepdendence Metropolis sampler. In this note, we show how to extend the result to obtain exact n-step transition probabilities for any n. This is done rst for a chain on a nite state space, and then...
متن کاملMulti-Rate Exact Discretization based on Diagonalization of a Linear System - A Multiple-Real-Eigenvalue Case
A multi-rate discrete-time model, whose response agrees exactly with that of a continuous-time original at all sampling instants for any sampling periods, is developed for a linear system, which is assumed to have multiple real eigenvalues. The sampling rates can be chosen arbitrarily and individually, so that their ratios can even be irrational. The state space model is obtained as a combinati...
متن کاملImportance Sampling for Continuous Time Bayesian Networks
A continuous time Bayesian network (CTBN) uses a structured representation to describe a dynamic system with a finite number of states which evolves in continuous time. Exact inference in a CTBN is often intractable as the state space of the dynamic system grows exponentially with the number of variables. In this paper, we first present an approximate inference algorithm based on importance sam...
متن کاملOn the pointfree counterpart of the local definition of classical continuous maps
The familiar classical result that a continuous map from a space $X$ to a space $Y$ can be defined by giving continuous maps $varphi_U: U to Y$ on each member $U$ of an open cover ${mathfrak C}$ of $X$ such that $varphi_Umid U cap V = varphi_V mid U cap V$ for all $U,V in {mathfrak C}$ was recently shown to have an exact analogue in pointfree topology, and the same was done for the familiar cla...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1998