An active set algorithm to estimate parameters in generalized linear models with ordered predictors

In biomedical studies, researchers are often interested in assessing the association between one or more ordinal explanatory variables and an outcome variable, at the same time adjusting for covariates of any type. The outcome variable may be continuous, binary, or represent censored survival times. In the absence of precise knowledge of the response function, using monotonicity constraints on the ordinal variables improves efficiency in estimating parameters, especially when sample sizes are small. An active set algorithm that can efficiently compute such estimators is proposed, and a characterization of the solution is provided. Having an efficient algorithm at hand is especially relevant when applying likelihood ratio tests in restricted generalized linear models, where one needs the value of the likelihood at the restricted maximizer. The algorithm is illustrated on a real life data set from oncology.

[1]  Mortaza Jamshidian,et al.  On Algorithms for Restricted Maximum Likelihood Estimation , 2002, Comput. Stat. Data Anal..

[2]  A FAST EM ALGORITHM FOR QUADRATIC OPTIMIZATION SUBJECT TO CONVEX CONSTRAINTS , 2007 .

[3]  A. Albert,et al.  On the existence of maximum likelihood estimates in logistic regression models , 1984 .

[4]  A. Shapiro.  Towards a unified theory of inequality constrained testing in multivariate analysis , 1988 .

[5]  B. Sen,et al.  Inconsistency of bootstrap: The Grenander estimator , 2010, 1010.3825.

[6]  Amy H Herring,et al.  Bayesian Inferences in the Cox Model for Order‐Restricted Hypotheses , 2003, Biometrics.

[7]  A. Fielding,et al.  Statistical Inference Under Order Restrictions. The Theory and Application of Isotonic Regression , 1974 .

[8]  Michael D. Perlman,et al.  One-Sided Testing Problems in Multivariate Analysis , 1969 .

[9]  Thomas J. Santner,et al.  A note on A. Albert and J. A. Anderson's conditions for the existence of maximum likelihood estimates in logistic regression models , 1986 .

[10]  L. Duembgen,et al.  Maximum likelihood estimation of a log-concave density and its distribution function: Basic properties and uniform consistency , 2007, 0709.0334.

[11]  J. Kalbfleisch.  Statistical Inference Under Order Restrictions , 1975 .

[12]  J. Wellner,et al.  The Support Reduction Algorithm for Computing Non‐Parametric Function Estimates in Mixture Models , 2008, Scandinavian journal of statistics, theory and applications.

[13]  F. T. Wright.  The Asymptotic Behavior of Monotone Regression Estimates , 1981 .

[14]  C. Holmes,et al.  Bayesian auxiliary variable models for binary and multinomial regression , 2006 .

[15]  Kaspar Rufibach,et al.  Active Set and EM Algorithms for Log-Concave Densities Based on Complete and Censored Data , 2007, 0707.4643.

[16]  J. Wellner,et al.  Estimation of a convex function: characterizations and asymptotic theory. , 2001 .

[17]  M. Silvapulle,et al.  On tests against one-sided hypotheses in some generalized linear models. , 1994, Biometrics.

[18]  D. Ghosh.  Incorporating monotonicity into the evaluation of a biomarker. , 2007, Biostatistics.

[19]  R. Fletcher.  Practical Methods of Optimization , 1988 .

[20]  Scott L. Zeger,et al.  Bootstrapping generalized linear models , 1991 .

[21]  R. Tibshirani,et al.  Generalized additive models for medical research , 1986, Statistical methods in medical research.

[22]  D.,et al.  Regression Models and Life-Tables , 2022 .

[23]  Adrian F. M. Smith,et al.  Bayesian Analysis of Constrained Parameter and Truncated Data Problems , 1991 .

[24]  Jon A Wellner,et al.  Computation of nonparametric convex hazard estimators via profile methods , 2009, Journal of nonparametric statistics.

[25]  I. Jolliffe,et al.  Nonlinear Multivariate Analysis , 1992 .

[26]  U. Grenander.  On the theory of mortality measurement , 1956 .

[27]  Fadoua Balabdaoui,et al.  Estimation of a k-monotone density , part 2 : algorithms for computation and numerical results , 2004 .

[28]  K. Rufibach.  Computing maximum likelihood estimators of a log-concave density function , 2007 .

[29]  Hari Mukerjee,et al.  Order-restricted inferences in linear regression , 1995 .

[30]  Lutz Dümbgen,et al.  Least squares and shrinkage estimation under bimonotonicity constraints , 2010, Stat. Comput..

[31]  Filippo Santambrogio,et al.  Least Squares estimation of two ordered monotone regression curves , 2009 .

[32]  Weizhen Wang,et al.  An Algorithm to Estimate Monotone Normal Means and its Application to Identify the Minimum Effective Dose , 2007, 0801.0079.

[33]  Daniel Taussky,et al.  Risk factors for developing a second upper aerodigestive cancer after radiotherapy with or without chemotherapy in patients with head-and-neck cancers: an exploratory outcomes analysis. , 2005, International journal of radiation oncology, biology, physics.

[34]  Federico Poloni.  Of Note , 2009 .

[35]  E. Steyerberg.  Clinical Prediction Models , 2008, Statistics for Biology and Health.

[36]  Willem J. Heiser,et al.  Local minima in categorical multiple regression , 2006, Comput. Stat. Data Anal..

[37]  Alex J Koning,et al.  Optimal Scaling of Interaction Effects in Generalized Linear Models , 2007, Multivariate behavioral research.

[38]  H. Barnett.  A Theory of Mortality , 1968 .

[39]  R. Dykstra,et al.  An Algorithm for Isotonic Regression for Two or More Independent Variables , 1982 .

[40]  G. Tutz,et al.  Penalized Regression with Ordinal Predictors , 2009 .

[41]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[42]  M. Silvapulle,et al.  Existence of Maximum Likelihood Estimates in Regression Models for Grouped and Ungrouped data , 1986 .

[43]  Roger Fletcher,et al.  Practical methods of optimization; (2nd ed.) , 1987 .

[44]  Lu Wang,et al.  Analysis on binary responses with ordered covariates and missing data , 2007, Statistics in medicine.

[45]  M. Kosorok.  Bootstrapping the Grenander estimator , 2008, 0805.2470.

[46]  L. Fahrmeir,et al.  Correction: Consistency and Asymptotic Normality of the Maximum Likelihood Estimator in Generalized Linear Models , 1985 .

[47]  Christian P. Robert,et al.  Maximum likelihood estimation under order restrictions by the prior feedback method , 1996 .

[48]  Brian Neelon,et al.  Bayesian Inference on Order‐Constrained Parameters in Generalized Linear Models , 2003, Biometrics.

[49]  Jean-Philippe Vial,et al.  Computing Maximum Likelihood Estimators of Convex Density Functions , 1998, SIAM J. Sci. Comput..

[50]  Peter Bacchetti,et al.  Additive Isotonic Models , 1989 .

[51]  Guang Cheng,et al.  Semiparametric additive isotonic regression , 2009 .

[52]  M. Hill,et al.  Nonlinear Multivariate Analysis. , 1990 .

[53]  F. T. Wright,et al.  Order restricted statistical inference , 1988 .

[54]  P. Diggle,et al.  Additive isotonic regression models in epidemiology. , 2000, Statistics in medicine.

[55]  J. Friedman,et al.  Estimating Optimal Transformations for Multiple Regression and Correlation. , 1985 .

[56]  Claudia Czado,et al.  Consistency and asymptotic normality of the maximum likelihood estimator in a zero-inflated generalized Poisson regression , 2005 .

[57]  Susan A. Murphy,et al.  Monographs on statistics and applied probability , 1990 .

[58]  L. Fahrmeir,et al.  Estimating and testing generalized linear models under inequality restrictions , 1994 .

[59]  P. Rousseeuw,et al.  Wiley Series in Probability and Mathematical Statistics , 2005 .

[60]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[61]  H. D. Brunk,et al.  Statistical inference under order restrictions : the theory and application of isotonic regression , 1973 .

[62]  Frank A. Wolak,et al.  An Exact Test for Multiple Inequality and Equality Constraints in the Linear Regression Model , 1987 .