Biometrika Advance Access originally published online on May 13, 2007
Biometrika 2007 94(2):496-501; doi:10.1093/biomet/asm029
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2007 Biometrika Trust
Miscellanea |
Identifiability of single-index models and additive-index models
Department of Mathematics, Ohio University, Athens, Ohio 45701, U.S.A.
Department of Mathematical Sciences, Clemson University, Clemson, South Carolina 29631, U.S.A.
linwei{at}math.ohiou.edu
kk{at}clemson.edu
Received for publication 1 March 2006. Accepted for publication 1 September 2006.
| Abstract |
|---|
We provide a proof for the identifiability for both single-index models and partially linear single-index models assuming only the continuity of the regression function, a condition much weaker than the differentiability conditions assumed in the existing literature. Our discussion is then extended to the identifiability of the additive-index models.
Key Words: Additive multiple-index model Identifiability Partially linear single-index model Single-index model