Anshul Singh

I have recently completed an M.Sc. in Mathematics, building a strong foundation in Statistics and Pure and Applied mathematics, following a B.Sc. (Honours) in Mathematics from the University of Delhi.

My Master’s thesis, supervised by Dr. Neeraj Joshi, focuses on the development of novel statistical methodologies to address the theoretical and inferential challenges arising from incomplete and degraded data. Part I, titled Multi-Component Stress–Strength Reliability under Middle Censoring Scheme, developed novel classical and Bayesian inferential procedures for estimating multicomponent stress–strength reliability from middle-censored data. This work has been published in the Journal of Statistical Computation and Simulation and can be viewed here. Part II extends this research by developing a unified inferential framework for multicomponent stress–strength reliability under the simultaneous presence of Middle Censoring and Numerical Masking, motivated by the reliability assessment of satellite telemetry and other modern engineering systems in which observations may be both incomplete and degraded. The study combines classical and Bayesian inference to develop a unified framework for reliability estimation and uncertainty quantification under such degraded observational conditions. This work is currently under review at a peer-reviewed journal.

My broader research interests include Bayesian statistics, missing data, uncertainty quantification, probabilistic machine learning and trustworthy AI, with particular interest in epistemic uncertainty, statistical learning theory and Imprecise Probability theory. I also have a strong technical interest in statistical computing, Monte carlo methos, hierarchical and latent-variable modelling in Python, R and C/C++. I am currently interested in pursuing Ph.D. research in Statistics and Statistical Machine Learning, particularly on problems that combine mathematical theory, statistical methodology and reliable AI.