Models for health disparity analysis

Qauntile forward regression for high-dimensional survival data (2023)

• Our main objective is to develop a quantile-specific prediction model using high-dimensional data. To demonstrate the effectiveness of this model, we present two examples. The first example focuses on the quantile-specific sequential selection of dietary factors associated with BMI. The second example highlights the quantile-specific sequential selection of risk factors for time-to-event analysis. Importantly, our model incorporates race and the interaction between race and risk factors to capture their impact on the predictions.
• paper        R code

Varying-coefficients for regional quantile via KNN-based LASSO with applications to health outcome study (2023)

• This novel framework for dynamic modeling of the associations between health outcomes and risk factors can capture the time-varying effects of age.
• paper        MATLAB code

A quantile regression decomposition estimation of disparities for complex survey data (2023+)

• The method proposes a quantile regression (QR) decomposition approach to disparity research using complex survey data.
• paper        R code

COVID-19 infectious disease modeling

Time-varying Poisson SIR model (2020)

• paper        R code

Survival data screening in high-dimensional data

Feature selection of ultrahigh-dimensional covariates with survival outcomes: a selective review (2017)

• paper

The Lq-norm learning for ultrahigh-dimensional survival data: an integrated framework (2018)

• The Lq-norm learning is proposed to detect predictors with various levels of impact, such as short- or long-term impact, on censored outcome.
• paper        R code

Integrated powered density (IPOD): screening ultrahigh dimensional covariates with survival outcome (2018)

• With a flexible weighting scheme, Kolmogorov statistic as a special case, IPOD method can detect early or late impact on censored outcome.
• paper        R code

Conditional screening for survival data (2018)

• The recently developed variable screening methods, though powerful in many practical setting, are less powerful in detecting marginally weak while jointly important signals. A new conditional screening method for survival outcome data computes the marginal contribution of each biomarker given priorly known biological information.
• paper        R code

Quantile adaptive model-free variable screening for high-dimensional heterogeneous data (2013)

• The proposed nonlinear independence screening procedure employs spline approximations to model the marginal effects at a quantile level of interest.
• paper        R code