TY  - JOUR
T1  - VARIABLE SELECTION WITH AKAIKE INFORMATION CRITERIA : A COMPARATIVE STUDY
AU  - Çetin, Meral Candan
AU  - Erar, Aydin
PY  - 2001
DA  - December
JF  - Hacettepe Journal of Mathematics and Statistics
PB  - Hacettepe University
WT  - DergiPark
SN  - 2651-477X
SP  - 89
EP  - 97
VL  - 31
LA  - en
AB  - In this paper, the problem of variable selection in linear regression isconsidered. This problem involves choosing the most appropriate modelfrom the candidate models. Variable selection criteria based on estimates ofthe Kullback-Leibler information are most common. Akaike's AIC and biascorrected AIC belong to this group of criteria. The reduction of the biasin estimating the Kullback-Leibler information can lead to better variableselection. In this study we have compared the Akaike Criterion based onFisher Information and AIC criteria based on Kullback-Leibler.
KW  - Akaike information criteria
KW  - robust selection
KW  - Kullback information
KW  - variable selection
CR  - .
.
UR  - https://dergipark.org.tr/en/pub/hujms/issue//530855
L1  - https://dergipark.org.tr/en/download/article-file/655391
ER  -