tag:blogger.com,1999:blog-2198942534740642384.post1777545666922105931..comments2023-10-24T03:16:41.009-07:00Comments on Econometrics Beat: Dave Giles' Blog: Concentrating, or Profiling, the Likelihood FunctionDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-2198942534740642384.post-38163088327898619172012-07-12T14:47:26.557-07:002012-07-12T14:47:26.557-07:00Brian - absolutely right - indeed this is probably...Brian - absolutely right - indeed this is probably the primary reason for profiling in the stats. literature. Thanks for pointing that out.Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-91089998928950961932012-07-12T14:44:22.444-07:002012-07-12T14:44:22.444-07:00Another reason for profiling a likelihood function...Another reason for profiling a likelihood function is to obtain confidence intervals for a subset of parameters when you don't think the usual normal-based asymptotic ones will have the right coverage. Here's a paper that describes the process: <br /><br />http://www.stata-journal.com/article.html?article=st0132Briannoreply@blogger.com