Bayes estimation of two-phase linear regression model (Q764451)

Summary: Let the regression model be \(Y_i = \beta_1 X_i + \epsilon_i\), where \(\epsilon_i\) are i.i.d. \(N (0, \sigma^2)\) random errors with variance \(\sigma^2 > 0\) but later it is found that there was a change in the system at some time point \(m\) and is reflected in the sequence after \(X_m\) by change in slope of the regression parameter \(\beta_2\). The problem of study is when and where this change had occurred. This is called the change point inference problem. The estimators of \(m,\;\beta_1,\;\beta_2\) are derived under asymmetric loss functions, namely Linex loss and eneral entropy loss functions. The effects of correct and wrong prior information on the Bayes estimates are studied.