Title :
Assessment of MCMC convergence: a time series and dynamical systems approach
Author :
Wolff, Rodney C. ; Nur, Darfiana ; Mengersen, Kerrie L.
Author_Institution :
Sch. of Math. Sci., Queensland Univ. of Technol., Brisbane, Qld., Australia
fDate :
6/23/1905 12:00:00 AM
Abstract :
Important in the application of Markov chain Monte Carlo (MCMC) methods is the determination that a search run has converged. Given that such searches typically take place in high-dimensional spaces, there are many pitfalls and difficulties in making such assessments. We discuss the use of phase randomisation as tool in the MCMC context, provide some details of its distributional properties for time series which enable its use as a convergence diagnostic, and contrast its performance with a selection of other widely used diagnostics. Some comments on analytical results, obtained via Edgeworth expansion, are also made
Keywords :
Markov processes; Monte Carlo methods; convergence of numerical methods; signal sampling; time series; Bayesian statistical methods; Edgeworth expansion; MCMC convergence; Markov chain Monte Carlo methods; convergence diagnostic; distributional properties; dynamical systems; high-dimensional spaces; phase randomisation; search convergence; time series resampling; Australia; Bayesian methods; Convergence; Density functional theory; Discrete Fourier transforms; Monte Carlo methods; Space technology; Statistical analysis; Tail; Testing;
Conference_Titel :
Statistical Signal Processing, 2001. Proceedings of the 11th IEEE Signal Processing Workshop on
Print_ISBN :
0-7803-7011-2
DOI :
10.1109/SSP.2001.955218
