UCL / rmcmc

Robust Markov chain Monte Carlo methods in R
http://github-pages.ucl.ac.uk/rmcmc/
Other
1 stars 0 forks source link

Provide per proposal defaults for target acceptance rate and initial scale #42

Closed matt-graham closed 1 month ago

matt-graham commented 1 month ago

Provides defaults for target_accept_prob and initial_scale parameter to scale adapters which are proposal (and for initial_scale dimension) dependent using 'optimal' values from Roberts and Rosenthal (2001)

and recommend values from Livingstone and Zanella (2022) for Barker proposal of target_accept_prob = 0.4 and initial_scale = 2.38 / (dimension)^(1/3).

codecov[bot] commented 1 month ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 100.00%. Comparing base (4e6a527) to head (547e3fd). Report is 1 commits behind head on main.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #42 +/- ## ========================================= Coverage 100.00% 100.00% ========================================= Files 10 10 Lines 378 391 +13 ========================================= + Hits 378 391 +13 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.