Research Interest:
Specialized Research Interests: Lifetime Data Analysis and Distribution Theory.
Main Research Interests: Likelihood-based Statistical Inference, Classical and Bayesian Inferences for Non-regular Family of Distributions, Asymptotic Theory, Accelerated Life-Testing, Competing Risks, Meta-Analysis of Lifetime Data, Optimal Design Criteria and Life-Testing Planning, and Optimal Bayesian Acceptance Sampling plan.
For more details, follow this link to my Research Statement
My research interests lie in likelihood-based statistical inference, classical and Bayesian inferences for non-regular families of distributions, optimal planning of life-testing experiments, optimal Bayesian acceptance sampling plans, accelerated life-testing, and meta-analysis of lifetime data. The applications of my research lie in reliability/industrial engineering and statistical quality control. Moreover, my background and passion allow me to extend my research into the field of actuarial science through collaborations.
I mainly work in three different research areas:
1. Classical and Bayesian inferences for a non-regular family of distributions: A non-regular family is a class of probability distributions that do not satisfy the regularity conditions typically assumed in classical statistical inference. Some key features of such a family of distributions are: Support depends on the parameter (e.g., uniform(0,θ), three-parameter gamma, etc.); its likelihood function may not be bounded for a certain range of parameter space, hence the maximum likelihood function does not exist; the likelihood function even may not be differentiable or integrable as needed, hence Fisher Information may not exist or be infinite. Moreover, standard results like MLE existence, consistency, and asymptotic normality may fail. Therefore, specialized or robust inferential techniques are needed, and I am interested in model selection and estimation for these models.
2. Bayesian acceptance sampling plan: In statistical quality control, an acceptance sampling plan (ASP) is a statistical rule/method used to decide whether to accept or reject a batch of manufactured products by testing only a sample from the batch, rather than checking every single item. Whereasa Bayesian acceptance sampling plan uses Bayesian decision theory to determine accept/reject rules based on prior knowledge and loss functions. It is useful in balancing inspection cost with quality assurance, and it helps in managing risks for both producers and consumers (false accept/reject cases). Its application lies in manufacturing (batteries, semiconductors, medical devices, etc.) and pharmaceuticals (drug batch testing), etc.
3. Optimal planning of the life-testing experiments: Lifetime testing, targeting at studying the lifetime distribution of products and forecasting of their expected lifetime, is of special interest in industrial engineering. In reliability analysis, optimal planning of life-testing experiments refers to designing a life-testing experiment to maximize information about lifetime parameters or minimize overall cost incurred during the test conduct under resource constraints.
At last, with my statistical expertise, I am always interested in exploring new research areas.