My research interests include interdisciplinary problems that require novel statistical solutions, especially those that could benefit from the use of Bayesian statistical formulations to incorporate expert inputs beyond observed historical data. My current interests are in Bayesian Methods, Spatial Modeling, Change Point Detection, Circular Statistics, Quantile Regression and Model Misspecification. Although I am interested in the applications of Data Science to various domains, my current focus is on problems in Crime Management and problems in Insurance. Motivated by practical problems, I also explore relevant theoretical questions that guide the appropriate use of statistical methods in practice and have made some contributions to the area of Bayesian statistical inference with Misspecified models and Quantile Regression. At IIMA, I teach courses related to Data Science tailored to the MBA students, PhD students and for participants in some of the Executive Education Programmes.
TOPICS/AREAS:
Data Science: Bayesian Statistics and Applications, Spatial Modeling, Change point detection, Quantile Regression.
Domains: Crime Management, Insurance, Bio-informatics.
PUBLICATIONS:
Kanti Mardia and Karthik Sriram (2022), Families of discrete circular distributions with some novel applications, Sankhya A. click here
Karthik Sriram and Peng Shi (2020), Stochastic loss reserving: A new perspective from a Dirichlet model, Journal of Risk and Insurance. [Awarded the Casualty Actuarial Society Best Paper Award by the American Risk and Insurance Association] click here
Dhruv Gupta and Karthik Sriram (2018), Impact of security expenditures in military alliances on violence from non-state actors: evidence from India. World Development, Vol 107, pg 338-357. click here
Kanti V Mardia, Karthik Sriram and Charlotte Deane (2018), A statistical model for helices with applications, Biometrics, Vol 74, No 3, pg 845-854. click here
Karthik Sriram and R.V. Ramamoorthi (2017), Correction to :"posterior consistency of Bayesian quantile regression based on the misspecified asymmetric Laplace density", Bayesian Analysis, Vol 12, No 4, pg 1217-1219. click here
Karthik Sriram, Peng Shi, and Pulak Ghosh (2016) , A Bayesian quantile regression model for insurance company costs data, Journal of Royal Statistical Society-Series A, Vol 179, Issue 1, pg 177-202. click here
Karthik Sriram, R.V. Ramamoorthi, and Pulak Ghosh (2016), On Bayesian Quantile Regression Using a Pseudo-joint Asymmetric Laplace Likelihood. Sankhya A, Vol 78, issue 1, pg 87-104
Karthik Sriram (2015), A sandwich likelihood correction for Bayesian quantile regression based on the misspecified asymmetric Laplace density, Statistics and Probability Letters, Vol 107, pg 18-26. click here
R.V. Ramamoorthi, Karthik Sriram, and Ryan Martin (2015), On posterior concentration in misspecified models, Bayesian Analysis, Vol 10, No. 4, pg 759-789. click here
Karthik Sriram, R.V. Ramamoorthi, and Pulak Ghosh (2013), Posterior consistency of Bayesian quantile regression based on the misspecified asymmetric Laplace density, Bayesian Analysis, Vol 8, No. 2, pg 479-504. click here
WORKING PAPERS
Karthik Sriram and R.V. Ramamoorthi: On posterior concentration rates for Bayesian quantile regression based on the misspecified asymmetric Laplace likelihood click here
Karthik Sriram, Dhruv Gupta, Rajiv Parikh: Movement of insurgent gangs: A Bayesian kernel density model for incomplete temporal data click here