Equivalence of predictors under real and over-parameterized linear models

Autor:	Shengjun Gan, Yongge Tian, Yuqin Sun
Rok vydání:	2016
Předmět:	Statistics and Probability 010104 statistics & probability Mathematical optimization Covariance matrix 010102 general mathematics Statistical inference Linear model Parameterized complexity 0101 mathematics Best linear unbiased prediction 01 natural sciences Equivalence (measure theory) Mathematics
Zdroj:	Communications in Statistics - Theory and Methods. 46:5368-5383
ISSN:	1532-415X 0361-0926
Popis:	Assume that a real linear model is over-parameterized as by adding some new regressors . In such a case, results of statistical inferences of the unknown parameters and under the two models are not necessarily the same. This paper aims at characterizing relationships between the best linear unbiased predictors (BLUPs) of the joint vector of the unknown parameters in the two models. In particular, we derive necessary and sufficient conditions for the BLUPs of to be equivalent under the real model and its over-parameterized counterpart.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::26035fe753ac392d6a91be3944fce322 https://doi.org/10.1080/03610926.2015.1100742 Zobrazit plný text záznamu