Regression Diagnostic Using Residual Plots

  • Oh, Kwang-Sik (Department of Informational Statistics, Catholic University of Daegu)
  • Published : 2001.08.01

Abstract

It is necessary to check the linearity of selected covariates in regression diagnostics. There are various graphical methods using residual plots such as partial residual plots, augmented partial residual plots and combining conditional expectation and residual plots. In this paper, we propose the modified pseudolikelihood ratio test statistics based on these residual plots to test linearity of selected covariate. These test statistics which measure the distance between the nonparametric and parametric models are derived as a ratio of quadratic forms. The approximate distribution of these statistics is calculated numerically by using three moments. The power comparison of these statistics is given.

Keywords

References

  1. Journal of The Royal Statistic Society B v.55 no.2 On the use of nonparametric regression for checking linear relationships Azzalini,A.;Bowman,A.W.
  2. Technometrics v.35 Exploring partial residual plots Cook.R.D.
  3. Technometrics v.38 Added-Variable Plots and Curvature in Linear Regression Cook.R.D.
  4. Residuals and influence in regression Cook,R.D.;Weisberg,S.
  5. Distributions in statistics:Continuous multivariate distributions Johnson,N.L.;Kotz,S.
  6. Technometrics v.14 The use of partial residual plots in regression analysis Larsen,W.A.;McCleary,S.J.
  7. Technometrics v.28 Augmented partial residual plots Mallows,C.L.
  8. Biometrika v.76 A cubic spline extension of the Durbin-Watson test Munson,P.J.;Jernigan,R.W.