A5034232594- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Economic and Environmental Valuation
- Health Systems, Economic Evaluations, Quality of Life
- Gestational Diabetes Research and Management
- Advanced Causal Inference Techniques
- Cancer Risks and Factors
- Laser and Thermal Forming Techniques
- Genetic and phenotypic traits in livestock
- Birth, Development, and Health
- Advanced Statistical Modeling Techniques
- Pregnancy and preeclampsia studies
- Nutritional Studies and Diet
- Advanced Statistical Methods and Models
- Neural Networks and Applications
- Decision-Making and Behavioral Economics
- Control Systems and Identification
- Nutrition, Genetics, and Disease
- Bayesian Methods and Mixture Models
- Sex and Gender in Healthcare
- Healthcare Policy and Management
- Statistical Distribution Estimation and Applications
- Genetic Associations and Epidemiology
- Advanced Measurement and Metrology Techniques
Yale University
2023-2025
University of Waterloo
2017-2020
ABSTRACT Major depressive disorder (MDD) is prevalent worldwide, substantially and negatively impacting both the quality length of life 280 million people globally. The genetic risk factors MDD have been studied in various previous research, but findings lack consistency. Sex/gender racial/ethnic disparities reported; however, many studies, represented by large‐scale genome‐wide association studies (GWASs) are known to diversity study cohorts. All Us a biorepository aiming focus on...
Suboptimal gestational weight gain (GWG), which is linked to increased risk of adverse outcomes for a pregnant woman and her infant, prevalent. In the study large cohort Canadian women, our goals are estimate individual growth trajectory using sparsely collected bodyweight data, identify factors affecting change during pregnancy, such as prepregnancy body mass index (BMI), dietary intakes physical activity. The first goal was achieved through functional principal component analysis (FPCA) by...
Background Health utilities from value sets for the EQ-5D-5L are commonly used in economic evaluations. We examined whether modeling spatial correlation among health states could improve precision of sets. Methods Using data 7 valuation studies, we compared predictive published linear model, a recently proposed cross-attribute level effects (CALE) and 2 Bayesian models with correlation. Predictive was quantified through root mean squared error (RMSE) out-of-sample predictions state-level on...
Since the sure independence screening (SIS) method by Fan and Lv, many different variable methods have been proposed based on measures under models. However, most of these are designed for specific In practice, we often very little information about data generating process can result in sets features. The heterogeneity presented here motivates us to combine various simultaneously. this paper, introduce a general ensemble-based framework efficiently results from multiple methods. consistency...
Abstract Two‐phase, response‐dependent sampling is often used in regression settings that involve expensive covariate measurements. Conditional maximum likelihood (CML) an attractive approach many cases as it avoids modelling of the distribution, unlike full likelihood. Scott & Wild (2011) introduced augmented CML which semi‐parametric efficient certain with a discrete response variable. We consider general models and show Scott–Wild estimator effects has same asymptotic efficiency two...
In eliciting utilities to value multiattribute utility instruments, discrete choice experiments (DCEs) administered online are less costly than interviewer-facilitated time tradeoff (TTO) tasks. DCEs capture on a latent scale and often coupled with small number of TTO tasks anchor the interval scale. Given nature data, design strategies that maximize set precision per response critical.Under simplifying assumptions, we expressed mean square prediction error (MSE) final as function J...
Two-phase outcome dependent sampling (ODS) is widely used in many fields, especially when certain covariates are expensive and/or difficult to measure. For two-phase ODS, the conditional maximum likelihood (CML) method very attractive because it can handle zero Phase 2 selection probabilities and avoids modeling covariate distribution. However, most existing CML-based methods use only sample thus may be less efficient than other methods. We propose a general empirical that uses CML augmented...
Abstract For regression with covariates missing not at random where the missingness depends on covariate values, complete‐case (CC) analysis leads to consistent estimation when is independent of response given all covariates, but it may have desired level efficiency. We propose a general empirical likelihood framework improve efficiency over CC analysis. expand methods in Bartlett et al. (2014, Biostatistics 15 , 719–730) and Xie Zhang (2017, Int J Biostat 13 1–20) that by modeling...
Two-phase outcome dependent sampling (ODS) is widely used in many fields, especially when certain covariates are expensive and/or difficult to measure. For two-phase ODS, the conditional maximum likelihood (CML) method very attractive because it can handle zero Phase 2 selection probabilities and avoids modeling covariate distribution. However, most existing CML-based methods use only sample thus may be less efficient than other methods. We propose a general empirical that uses CML augmented...