A Dynamic Graphical Method for Transformations and Curvature Specifications in Regression

  • Seo, Han-Son ;
  • Yoon, Min
  • Published : 2009.02.28


A dynamic graphical procedure is suggested to estimate optimal response transformation parameter and a curvature function of covariates in the regression model. Augmented partial residual plot is chosen for specifying a curvature. The proposed method is compared with a different approach (Soo, 2007) and is investigated efficiency by applying it to the real and the artificial data. The method is also extended to the 3D graphical situations.


Augmented partial residual plot;dynamic graphics;response transformation


  1. Berk, K. N. and Booth D. E. (1995). Seeing a curve in multiple regression, Technometrics, 37, 385-398
  2. Chambers, J. M., Cleveland, W. S., Kleiner, B. and Tukey, P. (1983). Graphical Methods for Data Analysis, Duxbury Press, Boston
  3. Cook, R. D. (1993). Exploring partial residual plots, Technometrics, 35, 351-362
  4. Cook, R. D. and Weisberg, S. (1994). Transforming a response variable for linearity, Biometrika, 81, 731-737
  5. Eaton, M. L. (1986). A characterization of spherical distributions, Journal of Multivariate Analysis, 20, 272-276
  6. Larsen, W. A. and Mccleary, S. J. (1972). The use of partial residual plots in regression analysis, Technometrics, 14, 781-790
  7. Mallows, C. L. (1986). Augmented partial residual plots, Technometrics, 28, 313-320
  8. Seo, H. S. (2007). A visual procedure for response transformations and curvature specifications, The 7th International Conference on Optimization: Techniques and Applications (ICOTA 7), CD-ROM
  9. Tierney, L. (1990). LISP-STAT: An Object-Oriented Environment for Statistical Computing and Dynamic Graphics, John Wiley & Sons, New York
  10. Weisberg, S. (1985). Applied Linear Regression, John Wiley & Sons, New York

Cited by

  1. Regression diagnostics for response transformations in a partial linear model vol.24, pp.1, 2013,