Reducing sensitivity to nuisance parameters in semiparametric models: a quasi-score method

SUMMARY This paper proposes a semiparametric extension of the projected score method of Waterman & Lindsay (1996) for the elimination of nuisance parameters. The procedure addresses cases where only the mean and the variance of the response variable are specified and where the mean function involves both parameters of interest and nuisance parameters. Important applications of the semiparametric model include quasilikelihood models for matched designs and for measurement error models (Carroll & Stefanski, 1990). As a result of the optimality and information-unbiasedness of the quasi-score function, a second-order quasi-score basis of estimating functions for the nuisance parameter is derived. Second-order locally ancillary estimating functions (Small & McLeish, 1994, pp. 81-4) are then obtained by solving a simple linear system that corresponds to a true projection for canonical exponential family distributions. Asymptotic arguments and simulation work show that the impact of nuisance parameters is considerably reduced when adopting the proposed approach.

[1]  P. McCullagh.  Quasi-Likelihood Functions , 1983 .

[2]  V. P. Godambe.  On ancillarityu and Fisher information in the presence of a nuisance parameter , 1984 .

[3]  D. A. Williams,et al.  Extra‐Binomial Variation in Logistic Linear Models , 1982 .

[4]  Christopher G. Small,et al.  Projection as a method for increasing sensitivity and eliminating nuisance parameters , 1989 .

[5]  P. McCullagh,et al.  Generalized Linear Models, 2nd Edn. , 1990 .

[6]  Bruce G. Lindsay,et al.  Projected score methods for approximating conditional scores , 1996 .

[7]  M. Crowder.  On linear and quadratic estimating functions , 1987 .

[8]  A. Desmond.  Optimal estimating functions, quasi-likelihood and statistical modelling , 1997 .

[9]  S. Zeger,et al.  Longitudinal data analysis using generalized linear models , 1986 .

[10]  J. Durbin.  Estimation of Parameters in Time‐Series Regression Models , 1960 .

[11]  Raymond J. Carroll,et al.  Conditional scores and optimal scores for generalized linear measurement-error models , 1987 .

[12]  V. P. Godambe.  Conditional likelihood and unconditional optimum estimating equations , 1976 .

[13]  V. P. Godambe.  An Optimum Property of Regular Maximum Likelihood Estimation , 1960 .

[14]  R. Royall.  Model robust confidence intervals using maximum likelihood estimators , 1986 .

[15]  J. Neyman,et al.  Consistent Estimates Based on Partially Consistent Observations , 1948 .

[16]  John A. Nelder,et al.  Generalized linear models. 2nd ed. , 1993 .

[17]  C. Small,et al.  Hilbert Space Methods in Probability and Statistical Inference , 1994 .

[18]  B. Lindsay.  Conditional score functions: Some optimality results , 1982 .

[19]  R. W. Wedderburn.  Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method , 1974 .

[20]  B. Lindsay.  Using Empirical Partially Bayes Inference for Increased Efficiency , 1985 .

[21]  V. P. Godambe.  On sufficiency and ancillarity in the presence of a nuisance parameter , 1980 .

[22]  Raymond J. Carroll,et al.  Approximate Quasi-likelihood Estimation in Models with Surrogate Predictors , 1990 .