Research Interests

Research Interests

Research Area : 

The "statistical science" realm constantly interacts with the ever-evolving challenges posed by the scientific community and industrial domains. A profound surge in data generation across various fields has brought statisticians to the forefront of deciphering this abundance of information. Their pivotal role entails extracting crucial patterns and trends, ultimately unraveling the true essence of what the data conveys. This process involves observing the pattern, developing models for capturing such phenomena, and then making inferences from the model. 

I am interested in various statistics and public health areas with real-life applications. My research involves developing statistical methodologies for "data-driven problems" from various applied disciplines (e.g., sample surveys, demography, epidemiology, clinical trials, survival experiments, reliability, and quality engineering). 

My primary research interest lies in modeling and drawing inferences from time-to-event data, censored data, masked data, longitudinal data, and competing risks.

My secondary research interests focus on public health, where I explored the development of individualized risk prediction models for severe asthma patients in real-world settings. In addition, I examined the causal relationship among the prognostic predictors for predicting severe asthma exacerbations using the Bayesian Network. Also, we were interested in comparing the causal pathway using randomized control trials and real-world data. In addition, I explored different ML and Bayesian methods to predict the progression of dementia using the Singaporean elderly population data. We identified essential risk factors of dementia and found that we could benefit by using longitudinal data over cross-sectional data to predict the outcome. 

Lastly, I am supporting the RWD/E team through comprehensive protocol and SAP reviews/development and statistical consultation for late-stage randomized controlled trials and non-interventional studies. I specialize in applying advanced causal inference methods to estimate treatment effects, account for confounding, and derive robust real-world evidence. This includes techniques such as inverse probability of treatment weighting (IPTW), marginal structural models (MSMs), g-methods, targeted maximum likelihood estimation (TMLE), trial emulation, estimands framework, and Bayesian borrowing methods.