USING PROFILE LIKELIHOOD FOR SEMIPARAMETRIC MODEL SELECTION WITH APPLICATION TO PROPORTIONAL HAZARDS MIXED MODELS.

We consider selection of nested and non-nested semiparametric models. Using profile likelihood we can define both a likelihood ratio statistic and an Akaike information for models with nuisance parameters. Asymptotic quadratic expansion of the log profile likelihood allows derivation of the asymptotic null distribution of the likelihood ratio statistic including the boundary cases, as well as unbiased estimation of the Akaike information by an Akaike information criterion. Our work was motivated by the proportional hazards mixed effects model (PHMM), which incorporates general random effects of arbitrary covariates and includes the frailty model as a special case. The asymptotic properties of its parameter estimate has recently been established, which enables the quadratic expansion of the log profile likelihood. For computation of the (profile) likelihood under PHMM we apply three algorithms: Laplace approximation, reciprocal importance sampling, and bridge sampling. We compare the three algorithms under different data structures, and apply the methods to a multi-center lung cancer clinical trial.

[1]  G. Fitzmaurice,et al.  Genetic and environmental contributions to the development of alcohol dependence in male twins. , 2004, Archives of general psychiatry.

[2]  Marie Davidian,et al.  A Monte Carlo EM algorithm for generalized linear mixed models with flexible random effects distribution. , 2002, Biostatistics.

[3]  Nicholas T. Longford.  Is ‘Which model . . .?’ the right question? , 2006 .

[4]  .. W. V. Der,et al.  On Profile Likelihood , 2000 .

[5]  F. Vaida,et al.  Conditional Akaike information for mixed-effects models , 2005 .

[6]  David V Glidden,et al.  Modelling clustered survival data from multicentre clinical trials , 2004, Statistics in medicine.

[7]  Runze Li,et al.  Variable selection for multivariate failure time data. , 2005, Biometrika.

[8]  M. van Glabbeke,et al.  Statistical methodology of phase III cancer clinical trials: advances and future perspectives. , 2002, European journal of cancer.

[9]  L. Tierney,et al.  Accurate Approximations for Posterior Moments and Marginal Densities , 1986 .

[10]  R. Gray.  Modeling Survival Data: Extending the Cox Model , 2002 .

[11]  Jianqing Fan,et al.  Profile likelihood inferences on semiparametric varying-coefficient partially linear models , 2005 .

[12]  C. Caroni,et al.  Graphical Tests for the Assumption of Gamma and Inverse Gaussian Frailty Distributions , 2005, Lifetime data analysis.

[13]  L. Wasserman,et al.  Computing Bayes Factors by Combining Simulation and Asymptotic Approximations , 1997 .

[14]  Ciprian M. Crainiceanu,et al.  Likelihood Ratio Testing for Zero Variance Components in Linear Mixed Models , 2008 .

[15]  Jonathan L. Blitstein,et al.  Design and analysis of group-randomized trials: a review of recent methodological developments. , 2004, American journal of public health.

[16]  W. Wong,et al.  Profile Likelihood and Conditionally Parametric Models , 1992 .

[17]  D. Stram,et al.  Variance components testing in the longitudinal mixed effects model. , 1994, Biometrics.

[18]  K. Liang,et al.  Asymptotic Properties of Maximum Likelihood Estimators and Likelihood Ratio Tests under Nonstandard Conditions , 1987 .

[19]  G. Verbeke,et al.  The effect of misspecifying the random-effects distribution in linear mixed models for longitudinal data , 1997 .

[20]  D. Commenges,et al.  Score test of homogeneity for survival data , 1995, Lifetime data analysis.

[21]  D. Ruppert,et al.  Likelihood ratio tests in linear mixed models with one variance component , 2003 .

[22]  David R. Anderson,et al.  Model selection and multimodel inference : a practical information-theoretic approach , 2003 .

[23]  D J Sargent,et al.  A general framework for random effects survival analysis in the Cox proportional hazards setting. , 1998, Biometrics.

[24]  Alan Agresti,et al.  Examples in which misspecification of a random effects distribution reduces efficiency, and possible remedies , 2004, Comput. Stat. Data Anal..

[25]  N. L. Johnson,et al.  Breakthroughs in Statistics , 1992 .

[26]  Q. Vuong.  Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses , 1989 .

[27]  H. Chernoff.  On the Distribution of the Likelihood Ratio , 1954 .

[28]  R. Prentice,et al.  Commentary on Andersen and Gill's "Cox's Regression Model for Counting Processes: A Large Sample Study" , 1982 .

[29]  Xiao-Li Meng,et al.  SIMULATING RATIOS OF NORMALIZING CONSTANTS VIA A SIMPLE IDENTITY: A THEORETICAL EXPLORATION , 1996 .

[30]  Jianqing Fan,et al.  Variable Selection for Cox's proportional Hazards Model and Frailty Model , 2002 .

[31]  José Cortiñas Abrahantes,et al.  Comparison of different estimation procedures for proportional hazards model with random effects , 2007, Comput. Stat. Data Anal..

[32]  G. Fitzmaurice,et al.  Genetic and environmental contributions to age of onset of alcohol dependence symptoms in male twins. , 2004, Addiction.

[33]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[34]  Susan A. Murphy,et al.  Asymptotic Theory for the Frailty Model , 1995 .

[35]  J Cai.  Hypothesis Testing of Hazard Ratio Parameters in Marginal Models for Multivariate Failure Time Data , 1999, Lifetime data analysis.

[36]  A. U.S.  An Asymptotic Theory for Model Selection Inference in General Semiparametric Problems , 2006 .

[37]  J. Deleeuw,et al.  Introduction to Akaike (1973) Information Theory and an Extension of the Maximum Likelihood Principle , 1992 .

[38]  F. Vaida,et al.  Conditional Akaike Information for Mixed Effects Models , 2004 .

[39]  Susan A Murphy,et al.  Two-level proportional hazards models. , 2002, Biometrics.

[40]  Erik T. Parner,et al.  Asymptotic theory for the correlated gamma-frailty model , 1998 .

[41]  F. Vaida,et al.  Proportional hazards model with random effects. , 2000, Statistics in medicine.

[42]  Nicholas T. Longford,et al.  Editorial: Model selection and efficiency—is ‘Which model …?’ the right question? , 2005 .

[43]  R. Gray.  Tests for Variation over Groups in Survival Data , 1995 .

[44]  Susan A. Murphy,et al.  Consistency in a Proportional Hazards Model Incorporating a Random Effect , 1994 .

[45]  Walter Zucchini,et al.  Model Selection , 2011, International Encyclopedia of Statistical Science.

[46]  H. T. V. Vu,et al.  Generalization of likelihood ratio tests under nonstandard conditions , 1997 .

[47]  Jianqing Fan,et al.  Generalized likelihood ratio statistics and Wilks phenomenon , 2001 .

[48]  E. L. Lehmann,et al.  Theory of point estimation , 1950 .

[49]  David V. Glidden.  Checking the adequacy of the gamma frailty model for multivariate failure times , 1999 .

[50]  A. Gelfand,et al.  Bayesian Model Choice: Asymptotics and Exact Calculations , 1994 .

[51]  John O'Quigley,et al.  Proportional hazards models with frailties and random effects , 2002, Statistics in medicine.

[52]  J. Kalbfleisch,et al.  The effects of mixture distribution misspecification when fitting mixed-effects logistic models , 1992 .

[53]  H C van Houwelingen,et al.  Time-dependent effects of fixed covariates in Cox regression. , 1995, Biometrics.

[54]  J. Palmgren,et al.  Estimation of Multivariate Frailty Models Using Penalized Partial Likelihood , 2000, Biometrics.

[55]  A. Manatunga,et al.  Diagnostic Plots for Assessing the Frailty Distribution in Multivariate Survival Data , 2001, Lifetime data analysis.