File:part1-fig-04-proposal-scale.png

Figure file · Markov chains, Gibbs sampling, and Metropolis–Hastings

Three random-walk Metropolis chains with the same \operatorname{N}(0,1) target and different proposal standard deviations s_q. The tiny proposal is accepted almost every time but advances slowly; the huge proposal repeats its current value through long runs of rejections. A moderate acceptance rate is useful only because it accompanies larger moves and much weaker autocorrelation. The displayed acceptance rate is a.

Three random-walk Metropolis chains with the same \operatorname{N}(0,1) target and different proposal standard deviations s_q. The tiny proposal is accepted almost every time but advances slowly; the huge proposal repeats its current value through long runs of rejections. A moderate acceptance rate is useful only because it accompanies larger moves and much weaker autocorrelation. The displayed acceptance rate is a.

Size of this preview: 786 × 443 pixels. Select it to open the high-resolution file.

Summary

Description. Three random-walk Metropolis chains with the same \operatorname{N}(0,1) target and different proposal standard deviations s_q. The tiny proposal is accepted almost every time but advances slowly; the huge proposal repeats its current value through long runs of rejections. A moderate acceptance rate is useful only because it accompanies larger moves and much weaker autocorrelation. The displayed acceptance rate is a.

Appears in. Markov chains, Gibbs sampling, and Metropolis–Hastings

Made by. A figure from the course chapter.

Download

File usage

These 3 pages show the file: