Liu-Type Logistic Estimator

Autor:	Birsen Eygi Erdogan, Deniz Inan
Rok vydání:	2013
Předmět:	Statistics and Probability Efficient estimator Minimum-variance unbiased estimator Bias of an estimator Mean squared error Modeling and Simulation Stein's unbiased risk estimate Statistics James–Stein estimator Econometrics Estimator Trimmed estimator Mathematics
Zdroj:	Communications in Statistics - Simulation and Computation. 42:1578-1586
ISSN:	1532-4141 0361-0918
DOI:	10.1080/03610918.2012.667480
Popis:	It is known that multicollinearity inflates the variance of the maximum likelihood estimator in logistic regression. Especially, if the primary interest is in the coefficients, the impact of collinearity can be very serious. To deal with collinearity, a ridge estimator was proposed by Schaefer et al. The primary interest of this article is to introduce a Liu-type estimator that had a smaller total mean squared error (MSE) than the Schaefer's ridge estimator under certain conditions. Simulation studies were conducted that evaluated the performance of this estimator. Furthermore, the proposed estimator was applied to a real-life dataset.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::28131c50dc192f51365043165c878aa2 https://doi.org/10.1080/03610918.2012.667480 Zobrazit plný text záznamu