Abstract
Case Deletion Diagnostics in Liu Semiparametric Regression Models
Highlights
Diagnostic techniques for the regression model have received great attention in statistical literature
In nonparametric and semiparametric regression models, diagnostic results are quite rare; among them one can refer to Eubank (1985), Silverman (1985), Thomas (1991), and Kim (1996) who studied the basic diagnostic building blocks such as the residuals, the leverage and the local influence for the choice of smoothing parameter
In general linear model Belsley et al (1989) noted that biased estimators are used to reduce the affect of multicollinearity and that the influence of some cases can be modified
Summary
Diagnostic techniques for the regression model have received great attention in statistical literature. Emami (2015) and Emami (2016) developed influence diagnostics based case deletion and local influence approach for ridge estima-tors and Liu estimators in semiparametric regression models, respectively. The goal of this paper is to supplement the work of Walker and Brich (1988) with such information and extend some ordinary linear regression influence diagnostics approach to LE in the linear semiparametric models. The multicollinearity is a problem when the primary interest is in the estimation of the parameters in a regression model. In the case of multicollinearity we know that when the correlation matrix has one or more small eigenvalues, the estimates of the regression coefficients can be large in absolute value.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
