Chenlei Leng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Chenlei Leng has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Binyan Jiang
Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University
4 shared publications Junlong Zhao
School of Statistics, Beijing Normal University
4 shared publicationsChao Liu
Beihang University People’s Republic of China
2 shared publications- Cheng Yong Tang
Department of Statistics, Operations, and Data Science Temple University
2 shared publications Hansheng Wang
Guanghua School of Management, Peking University, Beijing, China
2 shared publicationsLu Niu
Beihang University People’s Republic of China
2 shared publications
Latest Publications
- 2 July 2026
- Published by Taylor & Francis in Journal of Business & Economic Statistics
- Vol. 44 (4) , 1-34
Covariance Function Estimation for High-Dimensional Functional Time Series with Dual Factor Structures
- 2 July 2026
- Published by Taylor & Francis in Journal of Business & Economic Statistics
- Vol. 44 (4) , 1-34
Covariance Function Estimation for High-Dimensional Functional Time Series with Dual Factor Structures
- Open Access
- 4 June 2026
- Published by Taylor & Francis in Journal of the American Statistical Association
- p. 1-14
Residual Importance Weighted Transfer Learning for High-dimensional Linear Regression
1 Statistical techniques are most challenging in applications with small sample sizes, especially when accompanied by a large number of variables. While penalized regression has become a well-estab...
- 4 June 2026
- Published by Taylor & Francis in Journal of the American Statistical Association
- p. 1-14
Open AccessResidual Importance Weighted Transfer Learning for High-dimensional Linear Regression
1 Statistical techniques are most challenging in applications with small sample sizes, especially when accompanied by a large number of variables. While penalized regression has become a well-estab...
- Open Access
- 18 March 2026
- Published by Elsevier in SSRN Electronic Journal
Network Propagation Regression: A General and Interpretable Framework for Network-Linked Data
Network linked data are increasingly common in the social, biological, and information sciences, yet classical regression methods often fail to account for dependencies induced by network structure. We propose a unified and interpretable framework for regression on network linked data– Network Propagation Regression (NPR)–that flexibly captures both direct and higher-order network effects across diverse outcome types. NPR enables simple and efficient estimation via ordinary least squares for continuous outcomes and standard generalized linear model routines for binary, categorical, and time-to-event data. We establish consistency and asymptotic normality of the estimators under weak conditions, and develop hypothesis tests for the effective order of network influence. Simulations demonstrate that NPR outperforms established methods, particularly under model misspecification. An application to social media sentiment analysis illustrates the practical utility and robustness of NPR. Our framework provides a principled and general solution for regression analysis in complex networks.
- 18 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessNetwork Propagation Regression: A General and Interpretable Framework for Network-Linked Data
Network linked data are increasingly common in the social, biological, and information sciences, yet classical regression methods often fail to account for dependencies induced by network structure. We propose a unified and interpretable framework for regression on network linked data– Network Propagation Regression (NPR)–that flexibly captures both direct and higher-order network effects across diverse outcome types. NPR enables simple and efficient estimation via ordinary least squares for continuous outcomes and standard generalized linear model routines for binary, categorical, and time-to-event data. We establish consistency and asymptotic normality of the estimators under weak conditions, and develop hypothesis tests for the effective order of network influence. Simulations demonstrate that NPR outperforms established methods, particularly under model misspecification. An application to social media sentiment analysis illustrates the practical utility and robustness of NPR. Our framework provides a principled and general solution for regression analysis in complex networks.
- Open Access
Regression Analysis of Reciprocity in Directed Networks
Reciprocity--the tendency of individuals to form mutual ties--is a fundamental structural feature of many directed networks. Despite its ubiquity, reciprocity remains insufficiently integrated into statistical network models, particularly in relation to covariate information. In this paper, we introduce the $R^{2}$-Model, a novel and flexible framework that explicitly models reciprocity while incorporating covariate effects. Built upon a generalized $p_1$ model, our framework accommodates both network sparsity and node heterogeneity, offering the most comprehensive parametrization of reciprocity to date--capturing not only its baseline level but also how it systematically varies with observed covariates. To address the challenges posed by high dimensionality and nuisance parameters, we develop a conditional likelihood estimator that isolates and consistently estimates the reciprocity effects. We establish its theoretical guarantees, including consistency, asymptotic normality, and minimax optimality under broad sparsity regimes. Extensive simulations and real-world applications demonstrate the $R^{2}$-Model's flexibility, interpretability, and strong finite-sample performance, highlighting its practical utility for uncovering covariate-driven patterns of reciprocity in directed networks.Open AccessRegression Analysis of Reciprocity in Directed Networks
Reciprocity--the tendency of individuals to form mutual ties--is a fundamental structural feature of many directed networks. Despite its ubiquity, reciprocity remains insufficiently integrated into statistical network models, particularly in relation to covariate information. In this paper, we introduce the $R^{2}$-Model, a novel and flexible framework that explicitly models reciprocity while incorporating covariate effects. Built upon a generalized $p_1$ model, our framework accommodates both network sparsity and node heterogeneity, offering the most comprehensive parametrization of reciprocity to date--capturing not only its baseline level but also how it systematically varies with observed covariates. To address the challenges posed by high dimensionality and nuisance parameters, we develop a conditional likelihood estimator that isolates and consistently estimates the reciprocity effects. We establish its theoretical guarantees, including consistency, asymptotic normality, and minimax optimality under broad sparsity regimes. Extensive simulations and real-world applications demonstrate the $R^{2}$-Model's flexibility, interpretability, and strong finite-sample performance, highlighting its practical utility for uncovering covariate-driven patterns of reciprocity in directed networks. - Open Access
Modelling Directed Networks with Reciprocity
Asymmetric relational data is increasingly prevalent across diverse fields, underscoring the need for directed network models to address the complex challenges posed by their unique structures. Unlike undirected models, directed models can capture reciprocity, the tendency of nodes to form mutual links. In this work, we address a fundamental question: what is the effective sample size for modeling reciprocity? We examine this by analyzing the Bernoulli model with reciprocity, allowing for varying sparsity levels between non-reciprocal and reciprocal effects. We then extend this framework to a model that incorporates node-specific heterogeneity and link-specific reciprocity using covariates. Our findings reveal intriguing interplays between non-reciprocal and reciprocal effects in sparse networks. We propose a straightforward inference procedure based on maximum likelihood estimation that operates without prior knowledge of sparsity levels, whether covariates are included or not.
Open AccessModelling Directed Networks with Reciprocity
Asymmetric relational data is increasingly prevalent across diverse fields, underscoring the need for directed network models to address the complex challenges posed by their unique structures. Unlike undirected models, directed models can capture reciprocity, the tendency of nodes to form mutual links. In this work, we address a fundamental question: what is the effective sample size for modeling reciprocity? We examine this by analyzing the Bernoulli model with reciprocity, allowing for varying sparsity levels between non-reciprocal and reciprocal effects. We then extend this framework to a model that incorporates node-specific heterogeneity and link-specific reciprocity using covariates. Our findings reveal intriguing interplays between non-reciprocal and reciprocal effects in sparse networks. We propose a straightforward inference procedure based on maximum likelihood estimation that operates without prior knowledge of sparsity levels, whether covariates are included or not.
- Open Access
- 19 November 2024
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (550) , 1281-1293
A Sparse Beta Regression Model for Network Analysis
- 19 November 2024
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (550) , 1281-1293
Open AccessA Sparse Beta Regression Model for Network Analysis
- Open Access
A two-way heterogeneity model for dynamic networks
Dynamic network data analysis requires joint modelling individual snapshots and time dynamics. This paper proposes a new two-way heterogeneity model towards this goal. The new model equips each node of the network with two heterogeneity parameters, one to characterize the propensity of forming ties with other nodes and the other to differentiate the tendency of retaining existing ties over time. Though the negative log-likelihood function is non-convex, it is locally convex in a neighbourhood of the true value of the parameter vector. By using a novel method of moments estimator as the initial value, the consistent local maximum likelihood estimator (MLE) can be obtained by a gradient descent algorithm. To establish the upper bound for the estimation error of the MLE, we derive a new uniform deviation bound, which is of independent interest. The usefulness of the model and the associated theory are further supported by extensive simulation and the analysis of some real network data sets.
Open AccessA two-way heterogeneity model for dynamic networks
Dynamic network data analysis requires joint modelling individual snapshots and time dynamics. This paper proposes a new two-way heterogeneity model towards this goal. The new model equips each node of the network with two heterogeneity parameters, one to characterize the propensity of forming ties with other nodes and the other to differentiate the tendency of retaining existing ties over time. Though the negative log-likelihood function is non-convex, it is locally convex in a neighbourhood of the true value of the parameter vector. By using a novel method of moments estimator as the initial value, the consistent local maximum likelihood estimator (MLE) can be obtained by a gradient descent algorithm. To establish the upper bound for the estimation error of the MLE, we derive a new uniform deviation bound, which is of independent interest. The usefulness of the model and the associated theory are further supported by extensive simulation and the analysis of some real network data sets.
- Open Access
Linear Discriminant Analysis with High-dimensional Mixed Variables
Datasets containing both categorical and continuous variables are frequently encountered in many areas, and with the rapid development of modern measurement technologies, the dimensions of these variables can be very high. Despite the recent progress made in modelling high-dimensional data for continuous variables, there is a scarcity of methods that can deal with a mixed set of variables. To fill this gap, this paper develops a novel approach for classifying high-dimensional observations with mixed variables. Our framework builds on a location model, in which the distributions of the continuous variables conditional on categorical ones are assumed Gaussian. We overcome the challenge of having to split data into exponentially many cells, or combinations of the categorical variables, by kernel smoothing, and provide new perspectives for its bandwidth choice to ensure an analogue of Bochner's Lemma, which is different to the usual bias-variance tradeoff. We show that the two sets of parameters in our model can be separately estimated and provide penalized likelihood for their estimation. Results on the estimation accuracy and the misclassification rates are established, and the competitive performance of the proposed classifier is illustrated by extensive simulation and real data studies.
Open AccessLinear Discriminant Analysis with High-dimensional Mixed Variables
Datasets containing both categorical and continuous variables are frequently encountered in many areas, and with the rapid development of modern measurement technologies, the dimensions of these variables can be very high. Despite the recent progress made in modelling high-dimensional data for continuous variables, there is a scarcity of methods that can deal with a mixed set of variables. To fill this gap, this paper develops a novel approach for classifying high-dimensional observations with mixed variables. Our framework builds on a location model, in which the distributions of the continuous variables conditional on categorical ones are assumed Gaussian. We overcome the challenge of having to split data into exponentially many cells, or combinations of the categorical variables, by kernel smoothing, and provide new perspectives for its bandwidth choice to ensure an analogue of Bochner's Lemma, which is different to the usual bias-variance tradeoff. We show that the two sets of parameters in our model can be separately estimated and provide penalized likelihood for their estimation. Results on the estimation accuracy and the misclassification rates are established, and the competitive performance of the proposed classifier is illustrated by extensive simulation and real data studies.
- Open Access
A Random Graph-based Autoregressive Model for Networked Time Series
Contemporary time series data often feature objects connected by a social network that naturally induces temporal dependence involving connected neighbours. The network vector autoregressive model is useful for describing the influence of linked neighbours, while recent generalizations aim to separate influence and homophily. Existing approaches, however, require either correct specification of a time series model or accurate estimation of a network model or both, and rely exclusively on least-squares for parameter estimation. This paper proposes a new autoregressive model incorporating a flexible form for latent variables used to depict homophily. We develop a first-order differencing method for the estimation of influence requiring only the influence part of the model to be correctly specified. When the part including homophily is correctly specified admitting a semiparametric form, we leverage and generalize the recent notion of neighbour smoothing for parameter estimation, bypassing the need to specify the generative mechanism of the network. We develop new theory to show that all the estimated parameters are consistent and asymptotically normal. The efficacy of our approach is confirmed via extensive simulations and an analysis of a social media dataset.
Open AccessA Random Graph-based Autoregressive Model for Networked Time Series
Contemporary time series data often feature objects connected by a social network that naturally induces temporal dependence involving connected neighbours. The network vector autoregressive model is useful for describing the influence of linked neighbours, while recent generalizations aim to separate influence and homophily. Existing approaches, however, require either correct specification of a time series model or accurate estimation of a network model or both, and rely exclusively on least-squares for parameter estimation. This paper proposes a new autoregressive model incorporating a flexible form for latent variables used to depict homophily. We develop a first-order differencing method for the estimation of influence requiring only the influence part of the model to be correctly specified. When the part including homophily is correctly specified admitting a semiparametric form, we leverage and generalize the recent notion of neighbour smoothing for parameter estimation, bypassing the need to specify the generative mechanism of the network. We develop new theory to show that all the estimated parameters are consistent and asymptotically normal. The efficacy of our approach is confirmed via extensive simulations and an analysis of a social media dataset.
- Open Access
Applied Regression Analysis of Correlations for Correlated Data
Correlated data are ubiquitous in today's data-driven society. While regression models for analyzing means and variances of responses of interest are relatively well-developed, the development of these models for analyzing the correlations is largely confined to longitudinal data, a special form of sequentially correlated data. This paper proposes a new method for the analysis of correlations to fully exploit the use of covariates for general correlated data. In a renewed analysis of the Classroom data, a highly unbalanced multilevel clustered data with within-class and within-school correlations, our method reveals informative insights on these structures not previously known. In another analysis of the malaria immune response data in Benin, a longitudinal study with time-dependent covariates where the exact times of the observations are not available, our approach again provides promising new results. At the heart of our approach is a new generalized z-transformation that converts correlation matrices constrained to be positive definite to vectors with unrestricted support, and is order-invariant. These two properties enable us to develop regression analysis incorporating covariates for the modelling of correlations via the use of maximum likelihood.
Open AccessApplied Regression Analysis of Correlations for Correlated Data
Correlated data are ubiquitous in today's data-driven society. While regression models for analyzing means and variances of responses of interest are relatively well-developed, the development of these models for analyzing the correlations is largely confined to longitudinal data, a special form of sequentially correlated data. This paper proposes a new method for the analysis of correlations to fully exploit the use of covariates for general correlated data. In a renewed analysis of the Classroom data, a highly unbalanced multilevel clustered data with within-class and within-school correlations, our method reveals informative insights on these structures not previously known. In another analysis of the malaria immune response data in Benin, a longitudinal study with time-dependent covariates where the exact times of the observations are not available, our approach again provides promising new results. At the heart of our approach is a new generalized z-transformation that converts correlation matrices constrained to be positive definite to vectors with unrestricted support, and is order-invariant. These two properties enable us to develop regression analysis incorporating covariates for the modelling of correlations via the use of maximum likelihood.