Bayes Estimation of Two-Phase Linear Regression Model. (27th July 2011)

• Record Type: Journal Article
• Authors: Pandya, Mayuri; Bhatt, Krishnam; Andharia, Paresh
• Other Names: Chin Kwai Sang Academic Editor.
• Is Part Of: International journal of quality, statistics, and reliability. Volume 2011(2011)
• Publication Date: 2011-07-27
• Subjects: Quality control -- Periodicals; Quality control; Electronic journals; Journals - full-text; Periodicals; 620.0045
• DOI: 10.1155/2011/357814
• Languages: English
• ISSNs: 1687-7144

Abstract

Abstract : Let the regression model beY i = β 1 X i + ε i, whereε i are i. i. d. N (0, σ 2 ) random errors with varianceσ 2 > 0 but later it was found that there was a change in the system at some point of timem and it is reflected in the sequence afterX m by change in slope, regression parameterβ 2 . The problem of study is when and where this change has started occurring. This is called change point inference problem. The estimators ofm, β 1, β 2 are derived under asymmetric loss functions, namely, Linex loss & General Entropy loss functions. The effects of correct and wrong prior information on the Bayes estimates are studied.