Smoothed Langevin proposals in Metropolis-Hastings algorithms

✍ Scribed by Øivind Skare; Fred Espen Benth; Arnoldo Frigessi

• Book ID: 104301444
• Publisher: Elsevier Science
• Year: 2000
• Tongue: English
• Weight: 256 KB
• Volume: 49
• Category: Article
• ISSN: 0167-7152
• DOI: 10.1016/s0167-7152(00)00067-5

✦ Synopsis

The Metropolis Adjusted Langevin Algorithm (MALA) samples from complex multivariate densities . The proposal density is based on a discretized version of a Langevin di usion, and is well deÿned only for continuously di erentiable densities . We propose a modiÿed MALA algorithm when this condition is not fulÿlled or when has very rapid variations. The algorithm is illustrated on the Strauss model, for which two di erent classes of smoothing are proposed.

In these examples smoothing gives advantages in terms of reduced asymptotic variance.