Summary
Fully parametric models, in particular the exponential and the Weibull, are widely used in reliability analysis where both the shape of the baseline hazard and the effect of a set of explanatory variables are of interest.
In order to compare the fitting of alternative parametric models we propose a graphical procedure based on log-odds and normal deviate residuals as diagnostic statistics. These residuals have been originally suggested for the purpose of outlier screening but their properties make them suitable for verifying assumptions on the distribution of the baseline hazard as well.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Baltazar-Aban, I. andPeÑa, E.A (1995), Properties of hazard-based residuals and implications in model diagnostics.Journal of the American Statistical Association, 90, 185–197.
Cox, D. R. (1961), Test of separate families of hypotheses.Proc. 4th Berkeley Symp., 1, 105–123.
Cox, D. R. andSnell, E. J. (1968), A general definition of residuals (with discussion).Journal of the Royal Statistical Society, B, 30, 248–275.
Crowley, J. andHu, M. (1977), Covariance analysis of heart transplant survival data.Journal of the American Statistical Association, 72, 27–36.
Gehan, E. A. andThomas, D. G. (1969), The performance of some two-sample tests in small samples with and without censoring.Biometrika, 56, 127–132.
Green, S. B., Lininger, L., Gail, M. H. andByar, D. P. (1979), Comparison of four tests foe equality of survival curves in the presence of stratification and censoring.Biometrika, 66, 417–428.
Kay, R. (1977), Proportional hazards regression models and the analysis of censored survival data.Applied Statistics, 26 (3), 227–237.
Lagakos, S. W. (1981), The graphical evaluation of explanatory variablesin proportinal hazard regression models.Biometrika, 68, 93–98.
Nardi, A. andSchemper, M. (1999), New residuals for cox regression and their application to outlier screening.Biometrics, 55, 523–529.
Puin, B. (1978), Test and efficiencies of separate regression models.Biometrika, 65 (2), 319–327.
Rubin, D. B. (1987),Multiple Imputation for Nonresponse in Surveys. New York: Wiley.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nardi, A. Comparing parametric models for reliability data via residual analysis. J. Ital. Statist. Soc. 9, 139–158 (2000). https://doi.org/10.1007/BF03178963
Issue Date:
DOI: https://doi.org/10.1007/BF03178963