Speedier convergence with adaptive metropolis hastings

This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location where you will have to reset your password.
Forum rules
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.

Speedier convergence with adaptive metropolis hastings

Postby JamieHall » Mon Mar 16, 2009 2:46 am

The attached file implements the Adaptive Metropolis Hastings algorithm of Haario, Saksman and Tamminen. Generally speaking, its performance is significantly better than the standard random-walk metropolis hastings when dealing with DSGEs, in the sense that it explores the posterior distribution more efficiently and more accurately. But of course, YMMV.

It is backwards-compatible with the standard Dynare implementation. To turn on the adaptive MH computation, include these lines before the "estimation" command:
Code: Select all
global amh_t0;
amh_t0 = 20000;

Reducing the value of amh_t0 will start the adaptation earlier, at the risk of mis-estimating the covariance matrix.

This code is based on some stand-alone work by Kristoffer Nimark and Francisco Barillas. Maximum respect to them; all errors are mine.
Attachments
random_walk_metropolis_hastings.m
(8.11 KiB) Downloaded 399 times
JamieHall
 
Posts: 20
Joined: Wed Feb 11, 2009 4:42 am

Return to Dynare contributions and examples

Who is online

Users browsing this forum: No registered users and 5 guests