LZLeilei Zeng

Leilei Zeng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Leilei Zeng has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Analyzing zero‐truncated recurrent event data by stratified regression with time‐varying coefficients

    by Anqi A. Chen, X. Joan Hu, Rhonda J. Rosychuk, Leilei Zeng

    This article presents a strategy for conducting regression analysis of zero‐truncated recurrent event data. The research is partly motivated by a pediatric mental health care (PMHC) program based on administrative data. We are particularly interested in how the occurrence of an event depends on its past occurrences and the associated covariates over time. We propose a stratified Cox regression model with time‐varying coefficients. An easy‐to‐implement procedure is provided for estimating the model parameters using the zero‐truncated data integrated with readily available population census information. We examine the proposed estimator's finite‐sample performance through simulation and establish its asymptotic properties. The PMHC program data are used throughout the article to motivate and illustrate the proposed approach.


  • Open Access

    Analyzing Zero-Truncated Recurrent Events by Stratified Regression with Time-Varying Coefficients

    by Anqi A. Chen, X. Joan Hu, Rhonda J. Rosychuk, Leilei Zeng

    This paper presents a strategy for analyzing zero-truncated recurrent events data. Motivated by a pediatric mental health care (PMHC) program, we are particularly concerned with how the event occurrence depends on the occurrences in the past. We consider a stratified Cox regression model with time-varying coefficients and propose a procedure for estimating the model parameters using the zero-truncated data integrated with population census information. We evaluate the finite-sample performance of the proposed estimator through simulation and establish its asymptotic properties. Data from the PMHC program are used throughout the paper to motivate and to illustrate the proposed approach.

  • Open Access

    Pseudo-empirical likelihood methods for causal inference

    by Jingyue Huang, Changbao Wu, Leilei Zeng

    Causal inference remains a key area of research for assessing treatment effects across various real-world settings. This paper introduces two inferential procedures for the average treatment effect (ATE) using a two-sample pseudo-empirical likelihood (PEL) approach. The first procedure uses estimated propensity scores for the formulation of the PEL function, yielding a maximum PEL estimator equivalent to the inverse probability weighted estimator. Our focus in this scenario is on the PEL ratio statistic and establishing its theoretical properties. The second procedure incorporates outcome regression models for doubly robust PEL inference. We establish the asymptotic distribution of the PEL ratio statistic and also propose a bootstrap method for constructing confidence intervals to bypass a scaling constant involved in the asymptotic distribution of the PEL ratio statistic that is very difficult to calculate. Simulation studies and an real data application illustrate the performance of our proposed methods with comparisons to existing ones.


  • Open Access

    A Finite Mixture Hidden Markov Model for Intermittently Observed Disease Process with Heterogeneity and Partially Known Disease Type

    by Yidan Shi, Leilei Zeng, Mary E. Thompson, Suzanne L. Tyas

    Continuous-time multistate models are widely used for analyzing interval-censored data on disease progression over time. Sometimes, diseases manifest differently and what appears to be a coherent collection of symptoms is the expression of multiple distinct disease subtypes. To address this complexity, we propose a mixture hidden Markov model, where the observation process encompasses states representing common symptomatic stages across these diseases, and each underlying process corresponds to a distinct disease subtype. Our method models both the overall and the type-specific disease incidence/prevalence accounting for sampling conditions and exactly observed death times. Additionally, it can utilize partially available disease-type information, which offers insights into the pathway through specific hidden states in the disease process, to aid in the estimation. We present both a frequentist and a Bayesian way to obtain the estimates. The finite sample performance is evaluated through simulation studies. We demonstrate our method using the Nun Study and model the development and progression of dementia, encompassing both Alzheimer's disease (AD) and non-AD dementia.

  • Open Access

    Sample Empirical Likelihood Methods for Causal Inference

    by Jingyue Huang, Changbao Wu, Leilei Zeng

    Causal inference is crucial for understanding the true impact of interventions, policies, or actions, enabling informed decision-making and providing insights into the underlying mechanisms that shape our world. In this paper, we establish a framework for the estimation and inference of average treatment effects using a two-sample empirical likelihood function. Two different approaches to incorporating propensity scores are developed. The first approach introduces propensity scores calibrated constraints in addition to the standard model-calibration constraints; the second approach uses the propensity scores to form weighted versions of the model-calibration constraints. The resulting estimators from both approaches are doubly robust. The limiting distributions of the two sample empirical likelihood ratio statistics are derived, facilitating the construction of confidence intervals and hypothesis tests for the average treatment effect. Bootstrap methods for constructing sample empirical likelihood ratio confidence intervals are also discussed for both approaches. Finite sample performances of the methods are investigated through simulation studies.

  • Open Access

    Pseudo-Empirical Likelihood Methods for Causal Inference

    by Jingyue Huang, Changbao Wu, Leilei Zeng

    Causal inference problems have remained an important research topic over the past several decades due to their general applicability in assessing a treatment effect in many different real-world settings. In this paper, we propose two inferential procedures on the average treatment effect (ATE) through a two-sample pseudo-empirical likelihood (PEL) approach. The first procedure uses the estimated propensity scores for the formulation of the PEL function, and the resulting maximum PEL estimator of the ATE is equivalent to the inverse probability weighted estimator discussed in the literature. Our focus in this scenario is on the PEL ratio statistic and establishing its theoretical properties. The second procedure incorporates outcome regression models for PEL inference through model-calibration constraints, and the resulting maximum PEL estimator of the ATE is doubly robust. Our main theoretical result in this case is the establishment of the asymptotic distribution of the PEL ratio statistic. We also propose a bootstrap method for constructing PEL ratio confidence intervals for the ATE to bypass the scaling constant which is involved in the asymptotic distribution of the PEL ratio statistic but is very difficult to calculate. Finite sample performances of our proposed methods with comparisons to existing ones are investigated through simulation studies.

  • Open Access

    Cognitive Reserve and Mild Cognitive Impairment

    by Maryam Iraniparast, Yidan Shi, Ying Wu, Leilei Zeng, Colleen J. Maxwell, Richard J. Kryscio, Philip D. St John, Karen S. SantaCruz, Suzanne L. Tyas

    Background and Objectives Little is known about the effect of education or other indicators of cognitive reserve on the rate of reversion from mild cognitive impairment (MCI) to normal cognition (NC) or the relative rate (RR) of reversion from MCI to NC vs progression from MCI to dementia. Our objectives were to (1) estimate transition rates from MCI to NC and dementia and (2) determine the effect of age, APOE, and indicators of cognitive reserve on the RR of reversion vs progression using multistate Markov modeling. Methods We estimated instantaneous transition rates between NC, MCI, and dementia after accounting for transition to death across up to 12 assessments in the Nun Study, a cohort study of religious sisters aged 75+ years. We estimated RRs of reversion vs progression for age, APOE, and potential cognitive reserve indicators: education, academic performance (high school grades), and written language skills (idea density, grammatical complexity). Results Of the 619 participants, 472 were assessed with MCI during the study period. Of these 472, 143 (30.3%) experienced at least one reverse transition to NC, and 120 of the 143 (83.9%) never developed dementia (mean follow-up = 8.6 years). In models adjusted for age group and APOE, higher levels of education more than doubled the RR ratio of reversion vs progression. Novel cognitive reserve indicators were significantly associated with a higher adjusted RR of reversion vs progression (higher vs lower levels for English grades: RR ratio = 1.83; idea density: RR ratio = 3.93; and grammatical complexity: RR ratio = 5.78). Discussion Knowledge of frequent reversion from MCI to NC may alleviate concerns of inevitable cognitive decline in those with MCI. Identification of characteristics predicting the rate of reversion from MCI to NC vs progression from MCI to dementia may guide population-level interventions targeting these characteristics to prevent or postpone MCI and dementia. Research on cognitive trajectories would benefit from incorporating predictors of reverse transitions and competing events, such as death, into statistical modeling. These results may inform the design and interpretation of MCI clinical trials, given that a substantial proportion of participants may experience improvement without intervention.


  • Augmented likelihood for incorporating auxiliary information into left-truncated data

    by Yidan Shi, Leilei Zeng, Mary E. Thompson, Suzanne L. Tyas

    Time-to-event data are often subject to left-truncation. Lack of consideration of the sampling condition will introduce bias and loss in efficiency of the estimation. While auxiliary information from the same or similar cohorts may be available, challenges arise due to the practical issue of accessibility of individual-level data and taking account of various sampling conditions for different cohorts. In this paper, we introduce a likelihood-based method to incorporate information from auxiliary data to eliminate the left-truncation problem and improve efficiency. A one-step Monte-Carlo Expectation-Maximization algorithm is developed to calculate an augmented likelihood through creating pseudo-data sets which extend the form and conditions of the observed sample. The method is illustrated by both a real dataset and simulation studies.


  • Mitigating bias from intermittent measurement of time‐dependent covariates in failure time analysis

    by Shu Jiang, Richard J. Cook, Leilei Zeng

    Cox regression models are routinely fitted to examine the association between time‐dependent markers and a failure time when analyzing data from clinical registries. Typically, the marker values are measured periodically at clinic visits with the recorded value carried forward until the next assessment. We examine the asymptotic behavior of estimators from Cox regression models under this observation and data handling scheme when the true relationship is based on a Cox model using the current value of the marker. Specifically, we explore the impact of the marker process dynamics, the clinic visit intensity, and the marginal failure rate on the limiting value of the estimator of the marker effect from the Cox model. We also illustrate how a joint multistate model that accommodates intermittent observation of the time‐varying marker can be formulated. Simulation studies demonstrate that the finite sample performance of the naive estimator aligns with the asymptotic results and shows good performance of the estimators from the joint model. We apply both methods to data from a study of bone markers and their effect on the development of skeletal complications in metastatic cancer.


  • Multistate analysis from cross‐sectional and auxiliary samples

    by Leilei Zeng, Richard J. Cook, Jooyoung Lee

    Epidemiological studies routinely involve cross‐sectional sampling of a population comprised of individuals progressing through life history processes. We consider features of a cross‐sectional sample in terms of the intensity functions of a progressive multistate disease process under stationarity assumptions. The limiting values of estimators for regression coefficients in naive logistic regression models are studied, and simulations confirm the key asymptotic results that are relevant in finite samples. We also consider the need for and the use of data from auxiliary samples, which enable one to fit the full multistate life history process. We conclude with an application to data from a national cross‐sectional sample assessing marker effects on psoriatic arthritis among individuals with psoriasis.