CERES Plot in Nonlinear Regression

  • Myung-Wook (Associate Professor, Department of Statistics, Sookmyung Women's University) ;
  • Hye-Wook (Lecturer Department of Statistics Sookmyung Women's University)
  • Published : 2000.04.01

Abstract

We explore the structure and usefulness of CERES plot as a basic tool for dealing with curvature as a function of the new predictor in nonlinear regression. If a predictor has a nonlinear effect and there are nonlinear relationships among the predictors the partial residual plot and augmented partial residual plot are not able to display the correct functional form of the predictor. Unlike these plots the CERES plot can show the correct from. In situations where nonlinearity exists in two predictors we extend the idea of CERES plot to three dimensions, This is illustrated by simulated data.

Keywords

References

  1. Plots, Transformations, and Regression Atkinson, A.C.
  2. Technometrics v.40 Regression diagnostic plots in 3-D Berk, K.N.
  3. Technometrics v.37 Seeing a curve in multiple regression Berk, K.N.;Booth, D.E.
  4. Sensitivity Analysis in Linear Regression Chatterjee, S.;Hadi, A.S.
  5. Technometrics v.35 Exploring partial residual plots Cook, R.D.
  6. Residuals and influence in regression Cook, R.D.;Weisberg, S.
  7. Technometrics v.31 Regression diagnostics with dynamic graphics Cook, R.D.;Weisberg, S.
  8. Journal of the American Statistical Association v.19 A method for handing curvilinear correlation for any number of variables Ezekiel, M.
  9. The Korean Communications in Statistics v.3 Partial residual plot in nonlinear regression Ahng, M.;Kim, J.
  10. Technometrics v.28 Augmented partial residuals Allows, C.L.