Limin Peng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Limin Peng has a total of 77 co-authors. Here are the top co-authors by number of shared publications:
- Amita K. Manatunga
Emory University
4 shared publications Amita Manatunga
Department of Biostatistics and Bioinformatics,Emory University
4 shared publicationsBo Wei
Department of Biostatistics and Bioinformatics, Emory University , Atlanta, Georgia , USA
4 shared publicationsHuijuan Ma
KLATASDS-MOE, School of Statistics and Academy of Statistics, and Interdisciplinary Sciences, East China Normal University
4 shared publications- Ying Guo
Department of Biostatistics and Bioinformatics, Emory University
4 shared publications HuiChuan J. Lai
Department of Nutritional Sciences, University of Wisconsin–Madison , Madison, Wisconsin , USA
3 shared publications
Latest Publications
- 3 July 2025
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 81 (3)
Exploring the heterogeneity in recurrent episode lengths based on quantile regression
Recurrent episode data frequently arise in chronic disease studies when an event of interest occurs repeatedly and each occurrence lasts for a random period of time. Understanding the heterogeneity in recurrent episode lengths can help guide dynamic and customized disease management. However, there has been relative sparse attention to methods tailored to this end. Existing approaches either do not confer direct interpretation on episode lengths or involve restrictive or unrealistic distributional assumptions, such as exchangeability of within-individual episode lengths. In this work, we propose a modeling strategy that overcomes these limitations through adopting quantile regression and sensibly incorporating time-dependent covariates. Treating recurrent episodes as clustered data, we develop an estimation procedure that properly handles the special complications, including dependent censoring, dependent truncation, and informative cluster size. Our estimation procedure is computationally simple and yields estimators with desirable asymptotic properties. Our numerical studies demonstrate the advantages of the proposed method over naive adaptations of existing approaches.
- 3 July 2025
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 81 (3)
Exploring the heterogeneity in recurrent episode lengths based on quantile regression
Recurrent episode data frequently arise in chronic disease studies when an event of interest occurs repeatedly and each occurrence lasts for a random period of time. Understanding the heterogeneity in recurrent episode lengths can help guide dynamic and customized disease management. However, there has been relative sparse attention to methods tailored to this end. Existing approaches either do not confer direct interpretation on episode lengths or involve restrictive or unrealistic distributional assumptions, such as exchangeability of within-individual episode lengths. In this work, we propose a modeling strategy that overcomes these limitations through adopting quantile regression and sensibly incorporating time-dependent covariates. Treating recurrent episodes as clustered data, we develop an estimation procedure that properly handles the special complications, including dependent censoring, dependent truncation, and informative cluster size. Our estimation procedure is computationally simple and yields estimators with desirable asymptotic properties. Our numerical studies demonstrate the advantages of the proposed method over naive adaptations of existing approaches.
- 21 April 2025
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (552) , 2535-2546
Dynamic Regression of Longitudinal Trajectory Features
Chronic disease studies often collect data on biological and clinical markers at follow-up visits to monitor disease progression. Viewing such longitudinal measurements governed by latent continuous trajectories, we develop a new dynamic regression framework to investigate the heterogeneity pattern of certain features of the latent individual trajectory that may carry substantive information on disease risk or status. Employing the strategy of multi-level modeling, we formulate the latent individual trajectory feature of interest through a flexible pseudo B-spline model with subject-specific random parameters, and then link it with the observed covariates through quantile regression, avoiding restrictive parametric distributional assumptions that are typically required by standard multi-level longitudinal models. We propose an estimation procedure from adapting the principle of conditional score and develop an efficient algorithm for implementation. Our proposals yield estimators with desirable asymptotic properties as well as good finite-sample performance as confirmed by extensive simulation studies. An application of the proposed method to a cohort of participants with mild cognitive impairment (MCI) in the Uniform Data Set (UDS) provides useful insights about the complex heterogeneous presentations of cognitive decline in MCI patients.
- 21 April 2025
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (552) , 2535-2546
Dynamic Regression of Longitudinal Trajectory Features
Chronic disease studies often collect data on biological and clinical markers at follow-up visits to monitor disease progression. Viewing such longitudinal measurements governed by latent continuous trajectories, we develop a new dynamic regression framework to investigate the heterogeneity pattern of certain features of the latent individual trajectory that may carry substantive information on disease risk or status. Employing the strategy of multi-level modeling, we formulate the latent individual trajectory feature of interest through a flexible pseudo B-spline model with subject-specific random parameters, and then link it with the observed covariates through quantile regression, avoiding restrictive parametric distributional assumptions that are typically required by standard multi-level longitudinal models. We propose an estimation procedure from adapting the principle of conditional score and develop an efficient algorithm for implementation. Our proposals yield estimators with desirable asymptotic properties as well as good finite-sample performance as confirmed by extensive simulation studies. An application of the proposed method to a cohort of participants with mild cognitive impairment (MCI) in the Uniform Data Set (UDS) provides useful insights about the complex heterogeneous presentations of cognitive decline in MCI patients.
- 23 December 2024
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (551) , 1724-1735
Partial Quantile Tensor Regression
Tensors, characterized as multidimensional arrays, are frequently encountered in modern scientific studies. Quantile regression has the unique capacity to explore how a tensor covariate influences different segments of the response distribution. In this work, we propose a partial quantile tensor regression (PQTR) framework, which novelly applies the core principle of the partial least squares technique to achieve effective dimension reduction for quantile regression with a tensor covariate. The proposed PQTR algorithm is computationally efficient and scalable to a large tensor covariate. Moreover, we uncover an appealing latent variable model representation for the PQTR algorithm, justifying a simple population interpretation of the resulting estimator. We further investigate the connection of the PQTR procedure with an envelope quantile tensor regression (EQTR) model, which defines a general set of sparsity conditions tailored to quantile tensor regression. We prove the root- consistency of the PQTR estimator under the EQTR model, and demonstrate its superior finite-sample performance compared to benchmark methods through simulation studies. We demonstrate the practical utility of the proposed method via an application to a neuroimaging study of post traumatic stress disorder (PTSD). Results derived from the proposed method are more neurobiologically meaningful and interpretable as compared to those from existing methods.- 23 December 2024
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 120 (551) , 1724-1735
Partial Quantile Tensor Regression
Tensors, characterized as multidimensional arrays, are frequently encountered in modern scientific studies. Quantile regression has the unique capacity to explore how a tensor covariate influences different segments of the response distribution. In this work, we propose a partial quantile tensor regression (PQTR) framework, which novelly applies the core principle of the partial least squares technique to achieve effective dimension reduction for quantile regression with a tensor covariate. The proposed PQTR algorithm is computationally efficient and scalable to a large tensor covariate. Moreover, we uncover an appealing latent variable model representation for the PQTR algorithm, justifying a simple population interpretation of the resulting estimator. We further investigate the connection of the PQTR procedure with an envelope quantile tensor regression (EQTR) model, which defines a general set of sparsity conditions tailored to quantile tensor regression. We prove the root- consistency of the PQTR estimator under the EQTR model, and demonstrate its superior finite-sample performance compared to benchmark methods through simulation studies. We demonstrate the practical utility of the proposed method via an application to a neuroimaging study of post traumatic stress disorder (PTSD). Results derived from the proposed method are more neurobiologically meaningful and interpretable as compared to those from existing methods.- 8 November 2024
- Published by SAGE Publications in Diabetes Technology & Therapeutics
- Vol. 26 (11) , 862-868
Continuous Glucose Monitoring Metrics and Hemoglobin A1c Relationship in Patients with Type 2 Diabetes Treated by Hemodialysis
Background: There is a need for accurate glycemic control metrics in patients with diabetes and end-stage kidney disease (ESKD). Hence, we assessed the relationship of continuous glucose monitoring (CGM) metrics and laboratory-measured hemoglobin A1c (HbA1c) in patients with type 2 diabetes (T2D) treated by hemodialysis. Methods: This prospective observational study included adults (age 18–80 years) with T2D (HbA1c 5%–12%), treated by hemodialysis (for at least 90 days). Participants used a Dexcom G6 Pro CGM for 10 days. Correlation analyses between CGM metrics [mean glucose, glucose management indicator (GMI), and time-in-range (TIR 70–180 mg/dL)] and HbA1c were performed. Results: Among 59 participants (mean age was 57.7 ± 9.3 years, 58% were female, 86% were non-Hispanic blacks), the CGM mean glucose level was 188.9 ± 45 mg/dL (95% CI: 177.2, 200.7), the mean HbA1c and GMI were 7.1% ± 1.3% and 7.8% ± 1.1%, respectively (difference 0.74% ± 0.95). GMI had a strong negative correlation with TIR 70–180 mg/dL (r = −0.96). The correlation between GMI and HbA1c (r = 0.68) was moderate. Up to 29% of participants had a discordance between HbA1c and GMI of 1%. Conclusions: In patients with diabetes and ESKD treated by hemodialysis, the GMI has a strong correlation with TIR, while HbA1c underestimated the average glucose and GMI. Given the limitations of HbA1c in this population, GMI or mean glucose and TIR may be considered as more appropriate glucose control markers.
- 8 November 2024
- Published by SAGE Publications in Diabetes Technology & Therapeutics
- Vol. 26 (11) , 862-868
Continuous Glucose Monitoring Metrics and Hemoglobin A1c Relationship in Patients with Type 2 Diabetes Treated by Hemodialysis
Background: There is a need for accurate glycemic control metrics in patients with diabetes and end-stage kidney disease (ESKD). Hence, we assessed the relationship of continuous glucose monitoring (CGM) metrics and laboratory-measured hemoglobin A1c (HbA1c) in patients with type 2 diabetes (T2D) treated by hemodialysis. Methods: This prospective observational study included adults (age 18–80 years) with T2D (HbA1c 5%–12%), treated by hemodialysis (for at least 90 days). Participants used a Dexcom G6 Pro CGM for 10 days. Correlation analyses between CGM metrics [mean glucose, glucose management indicator (GMI), and time-in-range (TIR 70–180 mg/dL)] and HbA1c were performed. Results: Among 59 participants (mean age was 57.7 ± 9.3 years, 58% were female, 86% were non-Hispanic blacks), the CGM mean glucose level was 188.9 ± 45 mg/dL (95% CI: 177.2, 200.7), the mean HbA1c and GMI were 7.1% ± 1.3% and 7.8% ± 1.1%, respectively (difference 0.74% ± 0.95). GMI had a strong negative correlation with TIR 70–180 mg/dL (r = −0.96). The correlation between GMI and HbA1c (r = 0.68) was moderate. Up to 29% of participants had a discordance between HbA1c and GMI of 1%. Conclusions: In patients with diabetes and ESKD treated by hemodialysis, the GMI has a strong correlation with TIR, while HbA1c underestimated the average glucose and GMI. Given the limitations of HbA1c in this population, GMI or mean glucose and TIR may be considered as more appropriate glucose control markers.
- Open Access
Time-In-Range Analyses of Functional Data Subject to Missing with Applications to Inpatient Continuous Glucose Monitoring
Continuous glucose monitoring (CGM) has been increasingly used in US hospitals for the care of patients with diabetes. Time in range (TIR), which measures the percent of time over a specified time window with glucose values within a target range, has served as a pivotal CGM-metric for assessing glycemic control. However, inpatient CGM is prone to a prevailing issue that a limited length of hospital stay can cause insufficient CGM sampling, leading to a scenario with functional data plagued by complex missingness. Current analyses of inpatient CGM studies, however, ignore this issue and typically compute the TIR as the proportion of available CGM glucose values in range. As shown by simulation studies, this can result in considerably biased estimation and inference, largely owing to the nonstationary nature of inpatient CGM trajectories. In this work, we develop a rigorous statistical framework that confers valid inference on TIR in realistic inpatient CGM settings. Our proposals utilize a novel probabilistic representation of TIR, which enables leveraging the technique of inverse probability weighting and semiparametric survival modeling to obtain unbiased estimators of mean TIR that properly account for incompletely observed CGM trajectories. We establish desirable asymptotic properties of the proposed estimators. Results from our numerical studies demonstrate good finite-sample performance of the proposed method as well as its advantages over existing approaches. The proposed method is generally applicable to other functional data settings similar to CGM.
Open AccessTime-In-Range Analyses of Functional Data Subject to Missing with Applications to Inpatient Continuous Glucose Monitoring
Continuous glucose monitoring (CGM) has been increasingly used in US hospitals for the care of patients with diabetes. Time in range (TIR), which measures the percent of time over a specified time window with glucose values within a target range, has served as a pivotal CGM-metric for assessing glycemic control. However, inpatient CGM is prone to a prevailing issue that a limited length of hospital stay can cause insufficient CGM sampling, leading to a scenario with functional data plagued by complex missingness. Current analyses of inpatient CGM studies, however, ignore this issue and typically compute the TIR as the proportion of available CGM glucose values in range. As shown by simulation studies, this can result in considerably biased estimation and inference, largely owing to the nonstationary nature of inpatient CGM trajectories. In this work, we develop a rigorous statistical framework that confers valid inference on TIR in realistic inpatient CGM settings. Our proposals utilize a novel probabilistic representation of TIR, which enables leveraging the technique of inverse probability weighting and semiparametric survival modeling to obtain unbiased estimators of mean TIR that properly account for incompletely observed CGM trajectories. We establish desirable asymptotic properties of the proposed estimators. Results from our numerical studies demonstrate good finite-sample performance of the proposed method as well as its advantages over existing approaches. The proposed method is generally applicable to other functional data settings similar to CGM.
- Open Access
- 1 September 2024
- Published by The R Foundation in The R Journal
- Vol. 16 (3) , 134-149
SLCARE: An R Package for Semiparametric Latent Class Analysis of Recurrent Events
- 1 September 2024
- Published by The R Foundation in The R Journal
- Vol. 16 (3) , 134-149
Open AccessSLCARE: An R Package for Semiparametric Latent Class Analysis of Recurrent Events
- 1 January 2024
- Published by International Press of Boston in Statistics and Its Interface
- Vol. 17 (1) , 79-90
Latent class proportional hazards regression with heterogeneous survival data
International Press of Boston - publishers of scholarly mathematical and scientific journals and books
- 1 January 2024
- Published by International Press of Boston in Statistics and Its Interface
- Vol. 17 (1) , 79-90
Latent class proportional hazards regression with heterogeneous survival data
International Press of Boston - publishers of scholarly mathematical and scientific journals and books
- 1 January 2023
- Published by Statistica Sinica (Institute of Statistical Science) in Statistica Sinica
- Vol. 33 (Spec ) , 1295-1318
Globally Adaptive Longitudinal Quantile Regression With High Dimensional Compositional Covariates
In this work, we propose a longitudinal quantile regression framework that enables a robust characterization of heterogeneous covariate-response associations in the presence of high-dimensional compositional covariates and repeated measurements of both the response and the covariates. We develop a globally adaptive penalization procedure that can consistently identify covariate sparsity patterns across a continuum set of quantile levels. The proposed estimation procedure properly aggregates longitudinal observations over time, and satisfies the sum-zero coefficient constraint needed for a proper interpretation of the effects of compositional covariates. We establish the oracle rate of the uniform convergence and weak convergence of the resulting estimators, and further justify the proposed uniform selector of the tuning parameter in terms of achieving global model selection consistency. We derive an efficient algorithm by incorporating existing R packages to facilitate stable and fast computation. Our extensive simulation studies confirm our theoretical findings. We apply the proposed method to a longitudinal study of cystic fibrosis children, where the associations between the gut microbiome and other diet-related biomarkers are of interest.
- 1 January 2023
- Published by Statistica Sinica (Institute of Statistical Science) in Statistica Sinica
- Vol. 33 (Spec ) , 1295-1318
Globally Adaptive Longitudinal Quantile Regression With High Dimensional Compositional Covariates
In this work, we propose a longitudinal quantile regression framework that enables a robust characterization of heterogeneous covariate-response associations in the presence of high-dimensional compositional covariates and repeated measurements of both the response and the covariates. We develop a globally adaptive penalization procedure that can consistently identify covariate sparsity patterns across a continuum set of quantile levels. The proposed estimation procedure properly aggregates longitudinal observations over time, and satisfies the sum-zero coefficient constraint needed for a proper interpretation of the effects of compositional covariates. We establish the oracle rate of the uniform convergence and weak convergence of the resulting estimators, and further justify the proposed uniform selector of the tuning parameter in terms of achieving global model selection consistency. We derive an efficient algorithm by incorporating existing R packages to facilitate stable and fast computation. Our extensive simulation studies confirm our theoretical findings. We apply the proposed method to a longitudinal study of cystic fibrosis children, where the associations between the gut microbiome and other diet-related biomarkers are of interest.
- 9 December 2022
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 79 (3) , 1947-1958
Tensor Response Quantile Regression with Neuroimaging Data
Collecting neuroimaging data in the form of tensors (i.e. multidimensional arrays) has become more common in mental health studies, driven by an increasing interest in studying the associations between neuroimaging phenotypes and clinical disease manifestation. Motivated by a neuroimaging study of post-traumatic stress disorder (PTSD) from the Grady Trauma Project, we study a tensor response quantile regression framework, which enables novel analyses that confer a detailed view of the potentially heterogeneous association between a neuroimaging phenotype and relevant clinical predictors. We adopt a sensible low-rank structure to represent the association of interest, and propose a simple two-step estimation procedure which is easy to implement with existing software. We provide rigorous theoretical justifications for the intuitive two-step procedure. Simulation studies demonstrate good performance of the proposed method with realistic sample sizes in neuroimaging studies. We conduct the proposed tensor response quantile regression analysis of the motivating PTSD study to investigate the association between fMRI resting-state functional connectivity and PTSD symptom severity. Our results uncover non-homogeneous effects of PTSD symptoms on brain functional connectivity, which cannot be captured by existing tensor response methods.
- 9 December 2022
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 79 (3) , 1947-1958
Tensor Response Quantile Regression with Neuroimaging Data
Collecting neuroimaging data in the form of tensors (i.e. multidimensional arrays) has become more common in mental health studies, driven by an increasing interest in studying the associations between neuroimaging phenotypes and clinical disease manifestation. Motivated by a neuroimaging study of post-traumatic stress disorder (PTSD) from the Grady Trauma Project, we study a tensor response quantile regression framework, which enables novel analyses that confer a detailed view of the potentially heterogeneous association between a neuroimaging phenotype and relevant clinical predictors. We adopt a sensible low-rank structure to represent the association of interest, and propose a simple two-step estimation procedure which is easy to implement with existing software. We provide rigorous theoretical justifications for the intuitive two-step procedure. Simulation studies demonstrate good performance of the proposed method with realistic sample sizes in neuroimaging studies. We conduct the proposed tensor response quantile regression analysis of the motivating PTSD study to investigate the association between fMRI resting-state functional connectivity and PTSD symptom severity. Our results uncover non-homogeneous effects of PTSD symptoms on brain functional connectivity, which cannot be captured by existing tensor response methods.
- 1 October 2022
- Vol. 69 (5) , 68-85
[Effects of Multicomponent Case Management on the Problem Behaviors of Patients With Dementia and Family Caregivers' Distress, Self-Efficacy, Depression, Burden and Health Promotion Behaviors].
Few of the interventions currently available for family caregivers (FCGs) of persons with dementia (PWDs) with long-term follow-ups have a grounding in theory and incorporate multicomponent case management formats. Based on Pearlin's Caregiving and Stress Process model, this study was developed to examine the effectiveness of a family-centered case management program for PWDs with early to moderate dementia in terms of reducing PWDs behavioral problems and improve FCG outcomes, including distress, self-efficacy, depression, caregiver burden, and health-promoting behaviors. This randomized, single-blind, parallel-controlled trial included 76 dyads of PWDs and their FCGs. The dyads were recruited from outpatient clinics at dementia centers in three district hospitals in northern Taiwan. The dyads were randomly assigned to the intervention group (IG, n = 39) and control group (CG, n = 37). The dyads in the IG received a four-month intervention with two home or clinic visits and two telephone interviews. The multi-component interventions provided assessment, education, consultations, support, and referrals to long-term care resources. The CG received routine care and two social phone calls. Data were collected upon enrollment (T0 = baseline) and at 4-,6-, and 12-months post-intervention (T1, T2, and T3, respectively). Generalized estimating equations were conducted to analyze the effects of the intervention. By controlling for the interaction between group and time, we made a comparison between IG and the CG. The results showed significant improvements from baseline measures in behavioral problems in the PWDs for mood, psychosis, and social engagement, and improvements in the FCGs for distress and self-efficacy for obtaining respite as well as for better control of distressing thoughts, feelings of depression, caregiver burden, and overall health promoting behaviors at T1 and T2 (p < 0.5). Significant improvements were also found in the IG for psychomotor regulation among PWDs and the self-efficacy of FCGs in managing the PWDs' disturbing behaviors and health promotion behaviors for nutrition at T1 (p < 0.5). There were no significant improvements in the outcome variables at T3. Significant interactions between group and time were found at the 6-month assessment (T2) for improvements in problem behaviors of PWDs and depression, caregiver burden, and distress in the FCGs. Positive effects on self-efficacy and health promotion behaviors among the FCGs were also achieved. The results suggest that a multicomponent case management intervention should be referenced in dementia care policymaking for FCGs and PWDs.
- 1 October 2022
- Vol. 69 (5) , 68-85
[Effects of Multicomponent Case Management on the Problem Behaviors of Patients With Dementia and Family Caregivers' Distress, Self-Efficacy, Depression, Burden and Health Promotion Behaviors].
Few of the interventions currently available for family caregivers (FCGs) of persons with dementia (PWDs) with long-term follow-ups have a grounding in theory and incorporate multicomponent case management formats. Based on Pearlin's Caregiving and Stress Process model, this study was developed to examine the effectiveness of a family-centered case management program for PWDs with early to moderate dementia in terms of reducing PWDs behavioral problems and improve FCG outcomes, including distress, self-efficacy, depression, caregiver burden, and health-promoting behaviors. This randomized, single-blind, parallel-controlled trial included 76 dyads of PWDs and their FCGs. The dyads were recruited from outpatient clinics at dementia centers in three district hospitals in northern Taiwan. The dyads were randomly assigned to the intervention group (IG, n = 39) and control group (CG, n = 37). The dyads in the IG received a four-month intervention with two home or clinic visits and two telephone interviews. The multi-component interventions provided assessment, education, consultations, support, and referrals to long-term care resources. The CG received routine care and two social phone calls. Data were collected upon enrollment (T0 = baseline) and at 4-,6-, and 12-months post-intervention (T1, T2, and T3, respectively). Generalized estimating equations were conducted to analyze the effects of the intervention. By controlling for the interaction between group and time, we made a comparison between IG and the CG. The results showed significant improvements from baseline measures in behavioral problems in the PWDs for mood, psychosis, and social engagement, and improvements in the FCGs for distress and self-efficacy for obtaining respite as well as for better control of distressing thoughts, feelings of depression, caregiver burden, and overall health promoting behaviors at T1 and T2 (p < 0.5). Significant improvements were also found in the IG for psychomotor regulation among PWDs and the self-efficacy of FCGs in managing the PWDs' disturbing behaviors and health promotion behaviors for nutrition at T1 (p < 0.5). There were no significant improvements in the outcome variables at T3. Significant interactions between group and time were found at the 6-month assessment (T2) for improvements in problem behaviors of PWDs and depression, caregiver burden, and distress in the FCGs. Positive effects on self-efficacy and health promotion behaviors among the FCGs were also achieved. The results suggest that a multicomponent case management intervention should be referenced in dementia care policymaking for FCGs and PWDs.