alt text 

Trambak Banerjee
PhD candidate in Statistics


Email: Trambak.Banerjee.2020[at]marshall.usc.edu
Address: USC Marshall School of Business
308 Bridge Hall
3670 Trousdale Parkway
Los Angeles, CA 90089-0809

I am a fourth year PhD candidate in Statistics co-advised by Prof. Gourab Mukherjee and Prof. Wenguang Sun in the department of Data Sciences and Operations at USC Marshall School of Business. I received MS degrees in Statistics and Mathematical Finance from the Indian Statistical Institute and University of Oxford, UK respectively, and a BS degree in Statistics from St. Xavier’s College, Kolkata. Prior to joining USC, I have served as a quantitative modeler at FICO and Evalueserve where I have developed statistical models for credit scoring and for estimating banking regulatory capital pertaining to Credit and Operational risk events.

Research Interests: Empirical Bayes and Shrinkage based methods, Sparse estimation theory, Predictive Inference, Joint modeling of Longitudinal and Time-to-Event data, Computational Statistics.

Publications & Articles in review

Software

  • casp - An R package for Coordinate-wise Adaptive Shrinkage Prediction in a high-dimensional non-exchangeable hierarchical Gaussian model with unknown location as well as unknown spiked covariance structure.

  • asus - An R package that implements the ASUS (Adpative SURE thresholding with Side Information) procedure for estimating a high-dimensional sparse parameter when along with the primary data we can also gather side information from secondary data sources.

  • fusionclust - An R package for clustering and feature screening in large scale problems. In particular, fusionclust provides the Big Merge Tracker (BMT) and COSCI algorithms for convex clustering and feature screening using an ℓ1 fusion penalty.

Teaching