Chenlei Leng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chenlei Leng has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Cheng Yong Tang
Department of Statistics and Applied Probability, National University of Singapore, Singapore 117546, Republic of Singapore
2 shared publications - Jinfeng Xu
City University of Hong Kong Department of Biostatistics Hong Kong, 999077
2 shared publications - Bo Li
School of Economics and Management, Tsinghua University, Beijing 100084, China
1 shared publication Cheng Yong Tang
1 shared publication- David John Nott
Department of Statistics and Applied Probability, National University of Singapore, Singapore
1 shared publication David Nott
1 shared publication
Latest Publications
- 17 October 2012
- Published by Springer Nature in Statistics and Computing
- Vol. 24 (1) , 123-136
Smoothing combined estimating equations in quantile regression for longitudinal data
- 17 October 2012
- Published by Springer Nature in Statistics and Computing
- Vol. 24 (1) , 123-136
Smoothing combined estimating equations in quantile regression for longitudinal data
- 16 May 2012
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 99 (3) , 703-716
Penalized empirical likelihood and growing dimensional general estimating equations
When a parametric likelihood function is not specified for a model, estimating equations may provide an instrument for statistical inference. Qin and Lawless (1994) illustrated that empirical likelihood makes optimal use of these equations in inferences for fixed low-dimensional unknown parameters. In this paper, we study empirical likelihood for general estimating equations with growing high dimensionality and propose a penalized empirical likelihood approach for parameter estimation and variable selection. We quantify the asymptotic properties of empirical likelihood and its penalized version, and show that penalized empirical likelihood has the oracle property. The performance of the proposed method is illustrated via simulated applications and a data analysis.
- 16 May 2012
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 99 (3) , 703-716
Penalized empirical likelihood and growing dimensional general estimating equations
When a parametric likelihood function is not specified for a model, estimating equations may provide an instrument for statistical inference. Qin and Lawless (1994) illustrated that empirical likelihood makes optimal use of these equations in inferences for fixed low-dimensional unknown parameters. In this paper, we study empirical likelihood for general estimating equations with growing high dimensionality and propose a penalized empirical likelihood approach for parameter estimation and variable selection. We quantify the asymptotic properties of empirical likelihood and its penalized version, and show that penalized empirical likelihood has the oracle property. The performance of the proposed method is illustrated via simulated applications and a data analysis.
- 18 October 2011
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 98 (4) , 1001-1006
Empirical likelihood and quantile regression in longitudinal data analysis
We propose a novel quantile regression approach for longitudinal data analysis which naturally incorporates auxiliary information from the conditional mean model to account for within-subject correlations. The efficiency gain is quantified theoretically and demonstrated empirically via simulation studies and the analysis of a real dataset.
- 18 October 2011
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 98 (4) , 1001-1006
Empirical likelihood and quantile regression in longitudinal data analysis
We propose a novel quantile regression approach for longitudinal data analysis which naturally incorporates auxiliary information from the conditional mean model to account for within-subject correlations. The efficiency gain is quantified theoretically and demonstrated empirically via simulation studies and the analysis of a real dataset.
- 30 September 2011
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 98 (4) , 821-830
Forward adaptive banding for estimating large covariance matrices
We propose a simple forward adaptive banding method for estimating large covariance matrices using the modified Cholesky decomposition. This approach requires the fitting of a prespecified set of models due to the adaptive banding structure and can be efficiently implemented. Aside from its computational attractiveness, we propose a novel Bayes information criterion that gives consistent model selection for estimating high dimensional covariance matrices. The method compares favourably to its competitors in simulation study.
- 30 September 2011
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 98 (4) , 821-830
Forward adaptive banding for estimating large covariance matrices
We propose a simple forward adaptive banding method for estimating large covariance matrices using the modified Cholesky decomposition. This approach requires the fitting of a prespecified set of models due to the adaptive banding structure and can be efficiently implemented. Aside from its computational attractiveness, we propose a novel Bayes information criterion that gives consistent model selection for estimating high dimensional covariance matrices. The method compares favourably to its competitors in simulation study.
- 5 April 2011
- Published by Wiley in Statistics in Medicine
- Vol. 30 (16) , 2015-2027
Parametric variable selection in generalized partially linear models with an application to assess condom use by HIV‐infected patients
To study significant predictors of condom use in HIV‐infected adults, we propose the use of generalized partially linear models and develop a variable selection procedure incorporating a least squares approximation. Local polynomial regression and spline smoothing techniques are used to estimate the baseline nonparametric function. The asymptotic normality of the resulting estimate is established. We further demonstrate that, with the proper choice of the penalty functions and the regularization parameter, the resulting estimate performs as well as an oracle procedure. Finite sample performance of the proposed inference procedure is assessed by Monte Carlo simulation studies. An application to assess condom use by HIV‐infected patients gains some interesting results, which cannot be obtained when an ordinary logistic model is used. Copyright © 2011 John Wiley & Sons, Ltd.
- 5 April 2011
- Published by Wiley in Statistics in Medicine
- Vol. 30 (16) , 2015-2027
Parametric variable selection in generalized partially linear models with an application to assess condom use by HIV‐infected patients
To study significant predictors of condom use in HIV‐infected adults, we propose the use of generalized partially linear models and develop a variable selection procedure incorporating a least squares approximation. Local polynomial regression and spline smoothing techniques are used to estimate the baseline nonparametric function. The asymptotic normality of the resulting estimate is established. We further demonstrate that, with the proper choice of the penalty functions and the regularization parameter, the resulting estimate performs as well as an oracle procedure. Finite sample performance of the proposed inference procedure is assessed by Monte Carlo simulation studies. An application to assess condom use by HIV‐infected patients gains some interesting results, which cannot be obtained when an ordinary logistic model is used. Copyright © 2011 John Wiley & Sons, Ltd.
- 1 December 2010
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 97 (4) , 905-920
Penalized high-dimensional empirical likelihood
We propose penalized empirical likelihood for parameter estimation and variable selection for problems with diverging numbers of parameters. Our results are demonstrated for estimating the mean vector in multivariate analysis and regression coefficients in linear models. By using an appropriate penalty function, we showthat penalized empirical likelihood has the oracle property. That is, with probability tending to 1, penalized empirical likelihood identifies the true model and estimates the nonzero coefficients as efficiently as if the sparsity of the true model was known in advance. The advantage of penalized empirical likelihood as a nonparametric likelihood approach is illustrated by testing hypotheses and constructing confidence regions. Numerical simulations confirm our theoretical findings.
- 1 December 2010
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 97 (4) , 905-920
Penalized high-dimensional empirical likelihood
We propose penalized empirical likelihood for parameter estimation and variable selection for problems with diverging numbers of parameters. Our results are demonstrated for estimating the mean vector in multivariate analysis and regression coefficients in linear models. By using an appropriate penalty function, we showthat penalized empirical likelihood has the oracle property. That is, with probability tending to 1, penalized empirical likelihood identifies the true model and estimates the nonzero coefficients as efficiently as if the sparsity of the true model was known in advance. The advantage of penalized empirical likelihood as a nonparametric likelihood approach is illustrated by testing hypotheses and constructing confidence regions. Numerical simulations confirm our theoretical findings.
- Open Access
Bayesian Adaptive Lasso
We propose the Bayesian adaptive Lasso (BaLasso) for variable selection and coefficient estimation in linear regression. The BaLasso is adaptive to the signal level by adopting different shrinkage for different coefficients. Furthermore, we provide a model selection machinery for the BaLasso by assessing the posterior conditional mode estimates, motivated by the hierarchical Bayesian interpretation of the Lasso. Our formulation also permits prediction using a model averaging strategy. We discuss other variants of this new approach and provide a unified framework for variable selection using flexible penalties. Empirical evidence of the attractiveness of the method is demonstrated via extensive simulation studies and data analysis.
Open AccessBayesian Adaptive Lasso
We propose the Bayesian adaptive Lasso (BaLasso) for variable selection and coefficient estimation in linear regression. The BaLasso is adaptive to the signal level by adopting different shrinkage for different coefficients. Furthermore, we provide a model selection machinery for the BaLasso by assessing the posterior conditional mode estimates, motivated by the hierarchical Bayesian interpretation of the Lasso. Our formulation also permits prediction using a model averaging strategy. We discuss other variants of this new approach and provide a unified framework for variable selection using flexible penalties. Empirical evidence of the attractiveness of the method is demonstrated via extensive simulation studies and data analysis.
- 14 September 2009
- Published by Taylor & Francis in Journal of Nonparametric Statistics
- Vol. 21 (7) , 863-876
Statistical inference for induced L-statistics: a random perturbation approach
Suppose that X and Y are two numerical characteristics defined for each individual in a population. In a random sample of (X, Y) with sample size n, denote the rth ordered X variate by X r:n and the associated Y variate, the induced rth order statistics, by Y [r:n], respectively. Induced order statistics arise naturally in the context of selection where individuals ought to be selected by their ranks in a related X value due to difficulty or high costs of obtaining Y at the time of selection. The induced L-statistics, which take the form of , are very useful in regression analysis, especially when the observations are subject to a type-II censoring scheme with respect to the dependent variable, or when the regression function at a given quantile of the predictor variable is of interest. The limiting variance of the induced L-statistics involve the underlying regression function and inferences based on nonparametric estimation are often unstable. In this paper, we consider the distributional approximation of the induced L-statistics by the random perturbation method. Large sample properties of the randomly perturbed induced L-statistics are established. Numerical studies are also conducted to illustrate the method and to assess its finite-sample performance.
- 14 September 2009
- Published by Taylor & Francis in Journal of Nonparametric Statistics
- Vol. 21 (7) , 863-876
Statistical inference for induced L-statistics: a random perturbation approach
Suppose that X and Y are two numerical characteristics defined for each individual in a population. In a random sample of (X, Y) with sample size n, denote the rth ordered X variate by X r:n and the associated Y variate, the induced rth order statistics, by Y [r:n], respectively. Induced order statistics arise naturally in the context of selection where individuals ought to be selected by their ranks in a related X value due to difficulty or high costs of obtaining Y at the time of selection. The induced L-statistics, which take the form of , are very useful in regression analysis, especially when the observations are subject to a type-II censoring scheme with respect to the dependent variable, or when the regression function at a given quantile of the predictor variable is of interest. The limiting variance of the induced L-statistics involve the underlying regression function and inferences based on nonparametric estimation are often unstable. In this paper, we consider the distributional approximation of the induced L-statistics by the random perturbation method. Large sample properties of the randomly perturbed induced L-statistics are established. Numerical studies are also conducted to illustrate the method and to assess its finite-sample performance.
- 24 April 2009
- Published by Springer Nature in Statistics and Computing
- Vol. 20 (2) , 165-176
Rank-based variable selection with censored data
A rank-based variable selection procedure is developed for the semiparametric accelerated failure time model with censored observations where the penalized likelihood (partial likelihood) method is not directly applicable. The new method penalizes the rank-based Gehan-type loss function with the ℓ 1 penalty. To correctly choose the tuning parameters, a novel likelihood-based χ 2-type criterion is proposed. Desirable properties of the estimator such as the oracle properties are established through the local quadratic expansion of the Gehan loss function. In particular, our method can be easily implemented by the standard linear programming packages and hence numerically convenient. Extensions to marginal models for multivariate failure time are also considered. The performance of the new procedure is assessed through extensive simulation studies and illustrated with two real examples.
- 24 April 2009
- Published by Springer Nature in Statistics and Computing
- Vol. 20 (2) , 165-176
Rank-based variable selection with censored data
A rank-based variable selection procedure is developed for the semiparametric accelerated failure time model with censored observations where the penalized likelihood (partial likelihood) method is not directly applicable. The new method penalizes the rank-based Gehan-type loss function with the ℓ 1 penalty. To correctly choose the tuning parameters, a novel likelihood-based χ 2-type criterion is proposed. Desirable properties of the estimator such as the oracle properties are established through the local quadratic expansion of the Gehan loss function. In particular, our method can be easily implemented by the standard linear programming packages and hence numerically convenient. Extensions to marginal models for multivariate failure time are also considered. The performance of the new procedure is assessed through extensive simulation studies and illustrated with two real examples.
- Open Access
Bayesian projection approaches to variable selection and exploring model uncertainty
A Bayesian approach to variable selection which is based on the expected Kullback-Leibler divergence between the full model and its projection onto a submodel has recently been suggested in the literature. Here we extend this idea by considering projections onto subspaces defined via some form of $L_1$ constraint on the parameter in the full model. This leads to Bayesian model selection approaches related to the lasso. In the posterior distribution of the projection there is positive probability that some components are exactly zero and the posterior distribution on the model space induced by the projection allows exploration of model uncertainty. We also consider use of the approach in structured variable selection problems such as ANOVA models where it is desired to incorporate main effects in the presence of interactions. Here we make use of projections related to the non-negative garotte which are able to respect the hierarchical constraints. We also prove a consistency result concerning the posterior distribution on the model induced by the projection, and show that for some projections related to the adaptive lasso and non-negative garotte the posterior distribution concentrates on the true model asymptotically.Open AccessBayesian projection approaches to variable selection and exploring model uncertainty
A Bayesian approach to variable selection which is based on the expected Kullback-Leibler divergence between the full model and its projection onto a submodel has recently been suggested in the literature. Here we extend this idea by considering projections onto subspaces defined via some form of $L_1$ constraint on the parameter in the full model. This leads to Bayesian model selection approaches related to the lasso. In the posterior distribution of the projection there is positive probability that some components are exactly zero and the posterior distribution on the model space induced by the projection allows exploration of model uncertainty. We also consider use of the approach in structured variable selection problems such as ANOVA models where it is desired to incorporate main effects in the presence of interactions. Here we make use of projections related to the non-negative garotte which are able to respect the hierarchical constraints. We also prove a consistency result concerning the posterior distribution on the model induced by the projection, and show that for some projections related to the adaptive lasso and non-negative garotte the posterior distribution concentrates on the true model asymptotically.