Abstract
We consider the standard linear multiple regression model in which the parameter of interest is the ratio of two regression coefficients. Our setup includes a broad range of applications. We show that the 1− α confidence interval for the interest parameter based on the profile, conditional profile, modified profile or adjusted profile likelihood can potentially become the entire real line, while appropriately chosen integrated likelihoods do not suffer from this drawback. We further explore the asymptotic length of confidence intervals in order to compare integrated likelihood-based proposals. The analysis is facilitated by an orthogonal parameterization.
Similar content being viewed by others
References
Barndorff-Nielsen O.E. (1983). On a formula for the distribution of the maximum likelihood estimator. Biometrika, 70: 345–365
Barndorff-Nielsen O.E. (1994). Adjusted versions of profile likelihood and directed likelihood, and extended likelihood. Journal of the Royal Statistical Society Series B, 56: 125–140
Berger J. O., Bernardo J.M. (1992). On the development of reference priors (with discussion). In: Bernardo J.M., Berger J.O., Dawid A.P., Smith A.F.M. (Eds), Bayesian Statistics Vol. 4. Oxford University Press, Oxford, pp. 35–60
Berger J.O., Liseo B., Wolpert R. (1999). Integrated likelihood methods for eliminating nuisance parameters (with discussion). Statistical Science, 14: 1–22
Cox D.R., Reid N. (1987). Parameter orthogonality and approximate conditional inference (with discussion). Journal of the Royal Statistical Society Series B, 49: 1–39
Creasy M.A. (1954). Limits for the ratio of the means. Journal of the Royal Statistical Society Series B, 16: 186–194
Datta G.S., DiCiccio T.J. (2001). On expected volumes of multidimensional confidence sets associated with the usual and adjusted likelihoods. Journal of the Royal Statistical Society Series B, 63: 691–703
DiCiccio T.J., Stern S.E. (1994). Frequentist and Bayesian Bartlett correction of test statistics based on adjusted profile likelihoods. Journal of the Royal Statistical Society Series B, 56: 397–408
Fieller E.C. (1954). Some problems in interval estimation. Journal of the Royal Statistical Society Series B, 16: 175–185
Finney D.J. (1978). Statistical method in biological assay. Charles Griffin and Company, London
Ghosh M., Yin M., Kim, Y-H. (2003). Objective Bayesian inference for ratios of regression coefficients in linear models. Statistica Sinica, 13: 409–422
Gleser L.J., Hwang J.T. (1987). The non-existence of 100(1-α)% confidence sets of finite expected diameters in error-in-variable and related models. The Annals of Statistics 15: 1351–1362
Kalbfleisch J.D., Sprott,D.A. (1970). Application of likelihood methods to models involving large number of parameters (with discussion). Journal of the Royal Statistical Society Series B, 32: 175–208
McCullagh P., Tibshirani R. (1990). A simple method for the adjustment of profile likelihoods. Journal of the Royal Statistical Society Series B, 52: 325–344
Mukerjee R., Reid N. (1999). On confidence intervals associated with the usual and adjusted likelihoods. Journal of the Royal Statistical Society Series B, 61: 945–954
Mukerjee R., Reid N. (2001). Comparison of test statistics via expected lengths of associated confidence intervals. Journal of Statistical Planning and Inference, 97: 141–151
Scheffé H. (1970). Multiple testing versus multiple estimation. Improper confidence sets. Estimation of directions and ratios. The Annals of Mathematical Statistics, 41: 1–29
Sweeting T.J. (1987). Discussion of paper by D. R. Cox and N. Reid. Journal of the Royal Statistical Society Series B 49: 20–21
Wallace D.L. (1958). Intersection region confidence procedures with an application to the location of the maximum in quadratic regression. The Annals of Mathematical Statistics, 29: 455–475
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Ghosh, M., Datta, G.S., Kim, D. et al. Likelihood-based Inference for the Ratios of Regression Coefficients in Linear Models. Ann Inst Stat Math 58, 457–473 (2006). https://doi.org/10.1007/s10463-005-0027-3
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-005-0027-3