The Annals of Statistics
Estimating the Dimension of a Model
✓ Free legal copy found
Published version, hosted by projecteuclid.org
This is the final version, as it appears in the journal.
Read the free PDF →Abstract
The problem of selecting one of a number of models of different dimensions is treated by finding its Bayes solution, and evaluating the leading terms of its asymptotic expansion. These terms are a valid large-sample criterion beyond the Bayesian context, since they do not depend on the a priori distribution.
DOI: 10.1214/aos/1176344136 · Publisher: Institute of Mathematical Statistics