Pitfalls of estimating the marginal likelihood using the modified harmonic mean

C-Tier
Journal: Economics Letters
Year: 2015
Volume: 131
Issue: C
Pages: 29-33

Score contribution per author:

0.505 = (α=2.02 / 2 authors) × 0.5x C-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

The modified harmonic mean is widely used for estimating the marginal likelihood. We investigate the empirical performance of two versions of this estimator: one based on the observed-data likelihood and the other on the complete-data likelihood. Through an empirical example using US and UK inflation, we show that the version based on the complete-data likelihood has a substantial bias and tends to select the wrong model, whereas the version based on the observed-data likelihood works well.

Technical Details

RePEc Handle
repec:eee:ecolet:v:131:y:2015:i:c:p:29-33
Journal Field
General
Author Count
2
Added to Database
2026-01-25