Measurement error in proportional hazards models for survival data with long-term survivors

This work studies a proportional hazards model for survival data with "long-term survivors",in which covariates are subject to linear measurement error.It is well known that the naive estimators from both partial and full likelihood methods are inconsistent under this measurement error model.For measurement error models,methods of unbiased estimating function and corrected likelihood have been proposed in the literature.In this paper,we apply the corrected partial and full likelihood approaches to estimate the model and obtain statistical inference from survival data with long-term survivors.The asymptotic properties of the estimators are established.Simulation results illustrate that the proposed approaches provide useful tools for the models considered.     

Parsimonious and Efficient Likelihood Composition by Gibbs Sampling

The traditional maximum likelihood estimator (MLE) is often of limited use in complex high-dimensional data due to the intractability of the underlying likelihood function. Maximum composite likelihood estimation (McLE) avoids full likelihood specification by combining a number of partial likelihood objects depending on small data subsets, thus enabling inference for complex data. A fundamental difficulty in making the McLE approach practicable is the selection from numerous candidate likelihood objects for constructing the composite likelihood function. In this article, we propose a flexible Gibbs sampling scheme for optimal selection of sub-likelihood components. The sampled composite likelihood functions are shown to converge to the one maximally informative on the unknown parameters in equilibrium, since sub-likelihood objects are chosen with probability depending on the variance of the corresponding McLE. A penalized version of our method generates sparse likelihoods with a relatively small number of components when the data complexity is intense. Our algorithms are illustrated through numerical examples on simulated data as well as real genotype single nucleotide polymorphism (SNP) data from a case–control study.     

非线性回归模型中的约束拟似然

在非线性回归模型中,拟得分函数是一类线性无偏估计函数中的最优者(GodambeandHeyde(1987),朱仲义(1996)),而由拟得分函数得到的拟似然估计在由线性无偏估计函数得到的估计类中具有渐近最优性(林路(1999)).本文则研究非线性回归模型中的有偏估计函数理论,构造了参数的约束拟似然估计,得到了约束拟似然的局部最优性,局部改进了拟似然估计,从而扩充了线性模型中的有偏估计理论.     

Partial size-and-shape distributions

The concepts of partial size-and-shape and partial shape are defined, with motivation from a study in human movement analysis. Some co-ordinates for partial shape for landmarks in three dimensions are given, and Gaussian models for the landmark co-ordinates are proposed. The main results involve the derivation of the partial size-and-shape distributions for the isotropic and general multivariate normal models for three-dimensional data. The partial shape distribution is given in the isotropic case. Maximum likelihood based inference is explored, and examples using simulated and real human movement data illustrate the methodology.     

基于经验似然的部分线性模型的统计诊断

经验似然方法己经被广泛应用于许多模型的统计推断.本文基于经验似然对部分线性模型进行统计诊断.首先给出模型的估计方程,进而得到模型参数的极大经验似然估计;其次,基于经验似然研究了三种不同的影响曲率;最后通过随机模拟和实例分析,说明了统计诊断方法的有效性.     

Inference on Clustered Survival Data Using Imputed Frailties

This article proposes a new method for fitting frailty models to clustered survival data that is intermediate between the fully parametric and nonparametric maximum likelihood estimation approaches. A parametric form is assumed for the baseline hazard, but only for the purpose of imputing the unobserved frailties. The regression coefficients are then estimated by solving an estimating equation that is the average of the partial likelihood score with respect to the conditional distribution of frailties given the observed data. We prove consistency and asymptotic normality of the resulting estimators and give associated closedform estimators of their variance. The algorithm is easy to implement and reduces to the ordinary Cox partial likelihood approach when the frailties have a degenerate distribution. Simulations indicate high efficiency and robustness of the resulting estimates. We apply our new approach to a study with clustered survival data on asthma in children in east Boston.     

Empirical likelihood for partial linear models

In this paper the empirical likelihood method due to Owen (1988,Biometrika,75, 237–249) is applied to partial linear random models. A nonparametric version of Wilks' theorem is derived. The theorem is then used to construct confidence regions of the parameter vector in the partial linear models, which has correct asymptotic coverage. A simulation study is conducted to compare the empirical likelihood and normal approximation based method. Research supported by NNSF of China and a grant to the first author for his excellent Ph.D. dissertation work in China. Research supported by Hong Kong RGC CERG No. HKUST6162/97P.     

m-目依误差下部分线性模型的经验似然统计推断

本文中,我们针对误差为m-相依序列的固定设计的部分线性模型,运用经验似然方法和分组经验似然方法,构造了回归参数的对数经验似然比检验统计量,并且证明了分组经验似然比检验统计量在参数取真值时是渐近地服从卡方分布的.模拟计算表明分组经验似然方法的有效性.     

Partial penalized empirical likelihood ratio test under sparse case

A consistent test via the partial penalized empirical likelihood approach for the parametric hypothesis testing under the sparse case, called the partial penalized empirical likelihood ratio (PPELR) test, is proposed in this paper. Our results are demonstrated for the mean vector in multivariate analysis and regression coefficients in linear models, respectively. And we establish its asymptotic distributions under the null hypothesis and the local alternatives of order n^?1/2 under regularity conditions. Meanwhile, the oracle property of the partial penalized empirical likelihood estimator also holds. The proposed PPELR test statistic performs as well as the ordinary empirical likelihood ratio test statistic and outperforms the full penalized empirical likelihood ratio test statistic in term of size and power when the null parameter is zero. Moreover, the proposed method obtains the variable selection as well as the p-values of testing. Numerical simulations and an analysis of Prostate Cancer data confirm our theoretical findings and demonstrate the promising performance of the proposed method in hypothesis testing and variable selection.     

Nonlinear regression modeling using regularized local likelihood method

We introduce a nonlinear regression modeling strategy, using a regularized local likelihood method. The local likelihood method is effective for analyzing data with complex structure. It might be, however, pointed out that the stability of the local likelihood estimator is not necessarily guaranteed in the case that the structure of system is quite complex. In order to overcome this difficulty, we propose a regularized local likelihood method with a polynomial function which unites local likelihood and regularization. A crucial issue in constructing nonlinear regression models is the choice of a smoothing parameter, the degree of polynomial and a regularization parameter. In order to evaluate models estimated by the regularized local likelihood method, we derive a model selection criterion from an information-theoretic point of view. Real data analysis and Monte Carlo experiments are conducted to examine the performance of our modeling strategy.     

Likelihood inferences for the link function without knowing the true underlying distributions

This article is concerned with inference about link function in generalized linear models. A parametric and yet robust likelihood approach is introduced to accomplish the intended goal. More specifically, it is demonstrated that one can convert normal and gamma likelihoods into robust likelihood functions for the link function. The asymptotic validity of the robust likelihood requires only the existence of the second moments of the underlying distributions. The application of this novel robust likelihood method is demonstrated on the Box–Cox transformation. Simulation studies and real data analysis are provided to demonstrate the efficacy of the new parametric robust procedures.     

Solving the Likelihood Equations

Given a model in algebraic statistics and data, the likelihood function is a rational function on a projective variety. Algebraic algorithms are presented for computing all critical points of this function, with the aim of identifying the local maxima in the probability simplex. Applications include models specified by rank conditions on matrices and the Jukes–Cantor models of phylogenetics. The maximum likelihood degree of a generic complete intersection is also determined.     

Estimating parameters in diffusion processes using an approximate maximum likelihood approach

We present an approximate Maximum Likelihood estimator for univariate Itô stochastic differential equations driven by Brownian motion, based on numerical calculation of the likelihood function. The transition probability density of a stochastic differential equation is given by the Kolmogorov forward equation, known as the Fokker-Planck equation. This partial differential equation can only be solved analytically for a limited number of models, which is the reason for applying numerical methods based on higher order finite differences.The approximate likelihood converges to the true likelihood, both theoretically and in our simulations, implying that the estimator has many nice properties. The estimator is evaluated on simulated data from the Cox-Ingersoll-Ross model and a non-linear extension of the Chan-Karolyi-Longstaff-Sanders model. The estimates are similar to the Maximum Likelihood estimates when these can be calculated and converge to the true Maximum Likelihood estimates as the accuracy of the numerical scheme is increased. The estimator is also compared to two benchmarks; a simulation-based estimator and a Crank-Nicholson scheme applied to the Fokker-Planck equation, and the proposed estimator is still competitive.     

Maximum likelihood inference for the Cox regression model with applications to missing covariates

In this paper, we carry out an in-depth theoretical investigation for existence of maximum likelihood estimates for the Cox model [D.R. Cox, Regression models and life tables (with discussion), Journal of the Royal Statistical Society, Series B 34 (1972) 187–220; D.R. Cox, Partial likelihood, Biometrika 62 (1975) 269–276] both in the full data setting as well as in the presence of missing covariate data. The main motivation for this work arises from missing data problems, where models can easily become difficult to estimate with certain missing data configurations or large missing data fractions. We establish necessary and sufficient conditions for existence of the maximum partial likelihood estimate (MPLE) for completely observed data (i.e., no missing data) settings as well as sufficient conditions for existence of the maximum likelihood estimate (MLE) for survival data with missing covariates via a profile likelihood method. Several theorems are given to establish these conditions. A real dataset from a cancer clinical trial is presented to further illustrate the proposed methodology.     

Optimizing information using the EM algorithm in item response theory

Latent trait models such as item response theory (IRT) hypothesize a functional relationship between an unobservable, or latent, variable and an observable outcome variable. In educational measurement, a discrete item response is usually the observable outcome variable, and the latent variable is associated with an examinee’s trait level (e.g., skill, proficiency). The link between the two variables is called an item response function. This function, defined by a set of item parameters, models the probability of observing a given item response, conditional on a specific trait level. Typically in a measurement setting, neither the item parameters nor the trait levels are known, and so must be estimated from the pattern of observed item responses. Although a maximum likelihood approach can be taken in estimating these parameters, it usually cannot be employed directly. Instead, a method of marginal maximum likelihood (MML) is utilized, via the expectation-maximization (EM) algorithm. Alternating between an expectation (E) step and a maximization (M) step, the EM algorithm assures that the marginal log likelihood function will not decrease after each EM cycle, and will converge to a local maximum. Interestingly, the negative of this marginal log likelihood function is equal to the relative entropy, or Kullback-Leibler divergence, between the conditional distribution of the latent variables given the observable variables and the joint likelihood of the latent and observable variables. With an unconstrained optimization for the M-step proposed here, the EM algorithm as minimization of Kullback-Leibler divergence admits the convergence results due to Csiszár and Tusnády (Statistics & Decisions, 1:205–237, 1984), a consequence of the binomial likelihood common to latent trait models with dichotomous response variables. For this unconstrained optimization, the EM algorithm converges to a global maximum of the marginal log likelihood function, yielding an information bound that permits a fixed point of reference against which models may be tested. A likelihood ratio test between marginal log likelihood functions obtained through constrained and unconstrained M-steps is provided as a means for testing models against this bound. Empirical examples demonstrate the approach.     

Asymptotic properties of pseudo maximum likelihood estimators and test in semi-parametric copula models with multiple change points

The main purpose of the present paper is to establish the asymptotic properties of pseudo maximum likelihood estimators of the parameters of a multiple change-point model in the multivariate copula models when marginal distributions are unspecified but the copula function is parametrized. A pseudo likelihood ratio-type statistic is proposed for testing a sequence of observations for no change in the copula parameter against possible changes. Finally, a weighted bootstrap procedure that aims at evaluating the limiting distributions is examined.     

Full likelihood inferences in the Cox model: an empirical likelihood approach

For the regression parameter β ₀ in the Cox model, there have been several estimators constructed based on various types of approximated likelihood, but none of them has demonstrated small-sample advantage over Cox’s partial likelihood estimator. In this article, we derive the full likelihood function for (β ₀, F ₀), where F ₀ is the baseline distribution in the Cox model. Using the empirical likelihood parameterization, we explicitly profile out nuisance parameter F ₀ to obtain the full-profile likelihood function for β ₀ and the maximum likelihood estimator (MLE) for (β ₀, F ₀). The relation between the MLE and Cox’s partial likelihood estimator for β ₀ is made clear by showing that Taylor’s expansion gives Cox’s partial likelihood estimating function as the leading term of the full-profile likelihood estimating function. We show that the log full-likelihood ratio has an asymptotic chi-squared distribution, while the simulation studies indicate that for small or moderate sample sizes, the MLE performs favorably over Cox’s partial likelihood estimator. In a real dataset example, our full likelihood ratio test and Cox’s partial likelihood ratio test lead to statistically different conclusions.     

Empirical likelihood for single-index models with responses missing at random

The missing response problem in single-index models is studied, and a bias-correction method to infer the index coefficients is developed. Two weighted empirical log-likelihood ratios with asymptotic chisquare are derived, and the corresponding empirical likelihood confidence regions for the index coefficients are constructed. In addition, the estimators of the index coefficients and the link function are defined, and their asymptotic normalities are proved. A simulation study is conducted to compare the empirical likelihood and the normal approximation based method in terms of coverage probabilities and average lengths of confidence intervals. A real example illustrates our methods.     

SOME ASYMPTOTIC INFERENCE IN QUASI-LIKELIHOOD NONLINEAR MODELS:A GEOMETRIC APPROACH

§ 1　IntroductionIt is well known that quasi-likelihood models introduced by Wedderburn[1 ] greatlywiden the scope of generalized linear models by using a much weaker assumption in whichonly the firstand second moments ofresponse vector Yare needed to replace the full distri-butional assumption about Y in the models.It has drawn considerable attention in recentliterature(e.g.see[2～ 6] and so on) .However,little work has been done on the issuefrom a geometric viewpoint.The purpose of this p…     

The confidence profile method: a Bayesian method for assessing health technologies.

The Confidence Profile Method is a Bayesian method for adjusting and combining pieces of evidence to estimate parameters, such as the effect of health technologies on health outcomes. The information in each piece of evidence is captured in a likelihood function that gives the likelihood of the observed results of the evidence as a function of possible values of the parameter. A posterior distribution is calculated from Bayes formula as the product of the likelihood function and a prior distribution. Multiple pieces of evidence are incorporated by successive applications of Bayes' formula. Pieces of evidence are adjusted for biases to internal or external validity by modeling the biases and deriving "adjusted" likelihood functions that incorporate the models. Likelihood functions have been derived for one-, two- and multi-arm prospective studies; 2 x 2, 2 x n and matched case-control studies, and cross-sectional studies. Biases that can be incorporated in likelihood functions include crossover in controlled trials, error in measurement outcomes, patient selection biases, differences in technologies, and differences in length of follow-up. Effect measures include differences of rates, ratios of rates, and odds ratios. The elements of the method are illustrated with an analysis of the effect of a thrombolytic agent on the difference in probability of 1-year survival after a heart attack.