Premium
A regression method for censored inverse‐Gaussian data
Author(s) -
Whitmore G. A.
Publication year - 1983
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.2307/3314888
Subject(s) - inverse gaussian distribution , mathematics , estimator , gaussian , statistics , regression , inverse , regression analysis , sliced inverse regression , censored regression model , computer science , distribution (mathematics) , mathematical analysis , physics , geometry , quantum mechanics
Multiple regression methods are considered for progressively right‐censored inverse‐Gaussian data. Maximum‐likelihood estimators are derived using the EM algorithm, and their asymptotic distributional properties are presented. The methodology is demonstrated using two case illustrations, one of which involves the defective feature of the inverse‐Gaussian distribution. The relationship of the methodology to regression methods for complete inverse‐Gaussian samples and to other regression methods for censored survival data is discussed.