Journal of Pharmacokinetics and Biopharmaceutics

, Volume 17, Issue 5, pp 601–614 | Cite as

Fitting heteroscedastic regression models to individual pharmacokinetic data using standard statistical software

  • David M. Giltinan
  • David Ruppert


In the analysis of individual pharmacokinetic data by nonlinear regression it is important to allow for possible heterogeneity of variance in the response. Two common methods of doing this are weighted least squares with appropriate weights or data transformation using a suitable transform. With either approach it is appealing to let the data determine the appropriate choice of weighting scheme or transformation. This article describes two methods of doing this which are easy to compute using standard statistical software. The first method is a generalized least squares scheme for the case where the variance is assumed proportional to an unknown power of the mean. The second involves applying a power transformation to both sides of the regression equation. It is shown that both techniques may be implemented using only nonlinear regression routines. Sample code is provided for their implementation using the SAS software package. However, the proposed methods are feasible using any software package that incorporates a nonlinear least squares routine, and are thus well suited to routine use.

Key words

generalized least squares weighted least squares transform both sides power transformations heteroscedasticity nonlinear regression 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    C. C. Peck, S. L. Beal, L. B. Sheiner, and A. I. Nichols. Extended least squares nonlinear regression: A possible solution to the “choice of weights” problem in analysis of individual pharmacokinetic data.J. Pharmacokin. Biopharm. 12:545–558 (1984).CrossRefGoogle Scholar
  2. 2.
    L. B. Sheiner and S. L. Beal. Pharmacokinetic parameter estimates from several least squares procedures: superiority of extended least squares.J. Pharmacokin. Biopharm. 13:185–201 (1985).CrossRefGoogle Scholar
  3. 3.
    G. E. P. Box and W. J. Hill. Correcting inhomogeneity of variance with power transformation weighting.Technometrics 16:385–389 (1974).CrossRefGoogle Scholar
  4. 4.
    D. J. Pritchard, J. Downie, and D. W. Bacon. Further consideration of heteroscedasticity in fitting kinetic models.Technometrics 19:227–236 (1977).CrossRefGoogle Scholar
  5. 5.
    N. P. Jewell and G. M. Raab. Difficulties in obtaining consistent estimators of variance parameters.Biometrika 68:221–226 (1981).CrossRefGoogle Scholar
  6. 6.
    D. J. Finney and P. Phillips. The form and estimation of a variance function with particular reference to radiommunoassay.Appl. Statist. 26:312–320 (1977).CrossRefGoogle Scholar
  7. 7.
    G. E. P. Box and D. R. Cox. An analysis of transformations.J. Roy. Statist. Soc. Series B 26:211–252 (1964).Google Scholar
  8. 8.
    G. E. P. Box, W. G. Hunter, and J. S. Hunter.Statistics for Experimenters., Wiley, New York, 1978, p. 234.Google Scholar
  9. 9.
    R. J. Carroll and D. Ruppert. Power transformations when fitting theoretical models to data.J. Am. Statist. Assoc. 79:321–328 (1984).CrossRefGoogle Scholar
  10. 10.
    R. J. Carroll and D. Ruppert. Diagnostics and robust estimation when transforming the regression model and the response.Technometrics 29:287–299 (1987).CrossRefGoogle Scholar
  11. 11.
    S. A. Kaplan, R. E. Weinfeld, C. W. Abruzzo, and M. Lewis. Pharmacokinetic profile of sulfisoxazole following intravenous, intramuscular, and oral administration in man.J. Pharm. Sci. 61:773–778 (1972).PubMedCrossRefGoogle Scholar
  12. 12.
    R. J. Carroll and D. Ruppert.Transformation and Weighting in Regression, Chapman and Hall, New York, 1988.CrossRefGoogle Scholar
  13. 13.
    W. M. Patefield. On the maximized likelihood function.Sankhyā B 39:92–96 (1971).Google Scholar
  14. 14.
    S. L. Beal and L. B. Sheiner. Heteroscedastic nonlinear regression.Technometrics 30:327–338 (1988).CrossRefGoogle Scholar
  15. 15.
    R. J. Carroll and D. Ruppert. A comparison between maximum likelihood and generalized least squares in a heteroscedastic linear model.J. Am. Statist. Assoc. 77:878–882 (1982).CrossRefGoogle Scholar
  16. 16.
    J. C. van Houwelingen. Use and abuse of variance models in regression.Biometrics 44:1073–1081 (1988).PubMedCrossRefGoogle Scholar
  17. 17.
    D. Ruppert, N. Cressie, and R. J. Carroll. A transformation/weighting model for estimating Michaelis-Menten parameters.Biometrics 45: 637–656 (1989).CrossRefGoogle Scholar

Copyright information

© Plenum Publishing Corporation 1989

Authors and Affiliations

  • David M. Giltinan
    • 1
  • David Ruppert
    • 2
  1. 1.Medical Research DivisionAmerican Cyanamid CompanyPearl River
  2. 2.School of Operations Research and Industrial EngineeringCornell UniversityIthaca

Personalised recommendations