Research Summary

I work in the field of Bayesian statistics, currently focusing on statistical network analysis. Earlier in my academic career I worked on copula models for longitudinal data. During my graduate studies, I was lead technical consultant to develop the methodology for Mean Years of Schooling, an indicator produced by the UNESCO Institute for Statistics, used in UNDP Human Development Index.

I am currently a postdoctoral fellow at Montreal Institute for Learning Algorithms (MILA).

My research interests are:

  • Bayesian nonparametrics
  • Statistical network analysis
  • Hierarchical models
  • High-dimensional statistics
  • Markov Chains
  • Statistical learning algorithms
  • Machine learning
  • Applied statistics

Papers in progress

  • Proactive disease surveillance through link prediction in global host-parasite networks. (joint work with Maxwell Farrell and others).

  • Travel time prediction using GPS data. (joint work with Laurent Charlin, Aurelie Labbe and Denis Larocque).

Manuscripts

  • Sub-clustering in decomposable graphs and size-varying junction trees. [arXiv]

  • On decomposable random graphs. [arXiv]

  • A hierarchical Bayesian model for predicting ecological interactions using evolutionary relationships, joint work with Maxwell Farrell and David Stephens. [arXiv]

Publications

  • Das, Kalyan, Mohamad Elmasri, and Arusharka Sen.
    A Skew‐normal copula‐driven GLMM. Statistica Neerlandica 70, no. 4 (2016): 396-413. [PDF].

Technical reports

  • A Model to Decompose Inclusions in Educational Attainment Tables.
    UNESCO Institute for Statistics. Internal report. August 2014.
    Technical report produced as a consultant for the Methodology Unit of the UIS in 2014.

  • A Model to Forecast Mean Years of Schooling Estimates.
    UNESCO Institute for Statistics. Internal report. August 2014.
    Technical report produced as a consultant for the Methodology Unit of the UIS in 2014.

  • UIS Methodology For The Estimation of Mean Years of Schooling.
    UNESCO Institute for Statistics. December 2013.
    Joint work with Friedrich Huebler and Brenda Tay-Lim. [PDF].

Thesis

  • PhD in Statistics. Thesis: On decomposable random graphs and link prediction models, McGill University. [link].

  • MSc in Mathematics. Thesis: A Skew-Normal Copula-Driven Generalized Linear Mixed Model For Longitudinal Data, Concordia University. [PDF].

Talks and posters

  • A hierarchical Bayesian model for predicting ecological interactions using evolutionary relationships (poster)
    Theoretical foundations for statistical network analysis. Isaac Newton Institute for Mathematical Sciences, Cambridge, UK August 2017.

  • Bayesian sampling of decomposable graphs (Conference poster)
    Statistical and computational challenges in networks and cybersecurity. University of Montreal. Montreal, QC. May 2015.

  • Bayesian link prediction model for host-parasite interactions
    Colloque panquébécois des étudiants de l’Institut des sciences mathématiques. University of Montreal. Montreal, QC. May 2015.
    Joint work with Maxwell Farrell.

  • Bipartite models for biology interactions
    Statistics-Biology exchange group seminar. McGill University. Montreal, QC. October 2013.

  • Mean years of schooling: New dataset and methodology
    Comparative International Education Society Annual Conference. Toronto, ON. March 2013.
    Joint work with Friedrich Huebler and Brenda Tay-Lim.

  • Skew-normal copulas
    8th World Congress in Probability and Statistics, Istanbul, Turkey. July 2012.
    Joint work with Kalyan Das and Arusharka Sen.