Job market paper 

Title: Fitting the probability Gamma difference model with real return data and application of large deviation.

Abstract: 

In this work, we applied our new technique to estimate the parameters of the Gamma Difference distribution, which is the distribution of the difference of two random variables, each having a Gamma distribution. In the gamma difference distribution, there are four parameters, and we introduced a novel method for estimating them in a previous paper. We verified them with real data on returns. We compare the performance of our process to that used by Klar (2014) on three types of datasets. Once we have a reasonable estimate of the parameters, we attempt to apply large deviation theory to estimate the tail probability of large losses over an extended period.

Methods of Large Deviations in Finance, Insurance and  its extension for real return analysis 

Publication in progress:

1. Title: Methods of Large Deviations in Finance,  Insurance, and their extension for real return analysis

Abstract: 

Financial Economics analyzes the risk to one's holdings by investigating the randomness in investment returns. In a portfolio context, this requires examining the correlations between the returns of the different components. Large deviation theory can help estimate the small probability of large losses, especially for Insurance Companies. Standard models use lognormal distributions of returns and Gaussian cupolas for correlations. Usually, when analyzing credit risk, researchers use Gaussian copulas to estimate the probability of large losses; in this paper, we have generalized this to Gamma copulas. Large Deviation theory is also helpful in generating samples from the tails of a distribution, because, by their very nature, they are rare. Importance sampling is a way of overcoming this problem. In this work, we describe briefly these methods. We analyze the Gamma Difference model and investigate the estimation of the parameters of their distributions, as well as their tail behavior. (Communicating Physica Journal)

Keywords: Importance sampling, large deviation, Gamma difference distribution model,  insurance model, Credit Risk model.

Presented at the Applied Economics Seminar at the Graduate Center, CUNY (Sep  2025). 

Presented at Applied Math Department Seminar at Fordham University (November  2025) 

2.  Title: A comparative study of vector autoregressive (VAR),  Bayesian vector autoregressive(BVAR), and Mixfrequency  Bayesian vector autoregressive (MBVAR)

Abstract: 

In modern macroeconomics, vector autoregressive (VAR) are standard tools widely exploited for structural analysis and forecasting. Here, we use the vector autoregressive (VAR) models in multivariate macroeconomic analysis. The Bayesian VARs are solid tools for forecasting. This paper employs classical and modern Bayesian vector autoregressive (BVAR) techniques for equal-frequency and mixed-frequency Bayesian VAR (MBVAR) models. The first part of the paper focused on classical analyses, including the unit root test, cointegration test, VECM analysis, and structural VAR analysis, using US GDP and other related data. Later, choosing the appropriate prior distribution, the same data was used for BVAR and MVAR analysis. Finally, the results (impulse responses, forecasting, and other exciting effects) were compared using VAR, BVAR, and MVAR analyses.(Communication Journal of Applied Econometrics) 

Working Papers:

Other Papers :

Title: A Comparative Clinical Study on Open Appendicectomy & Laparoscopic Appendicectomy 

Abstract: Abstract-Appendicectomy is one of the most commonly performed surgeries worldwide. As it was commonly done by conventional open method, but with the gradual advancement of more and more laparoscopic surgical procedures, laparoscopic appendectomy is also practiced nowadays.                      

Methods: A Total of 86 patients were initially selected for the study, and 6 patients were subsequently excluded. Out of 80 patients, 40 were in the open group and 40 were in the laparoscopy group. 

  Index term--Abscess, After8,  After16,  After24,  After 48, Collection, Collection Problem, Ceal.Leak, Cosmesis Scar, Delay Healing, Disruption, DtOTTime, Gamma distribution, G.I.Obstruction, IleusPost, Indoor Work, Likelihood, Lognormal distribution, LNGD,   Or,  Outdoor Work, Retention, Stay Days, Surgery, Transition Matrix. UTI, Vomiting, Weibull distribution, WoundAbcess.