Campuses:

Hossein Keshavarz

  • Industrial Postdoctoral Fellow
Location: 
436 Lind Hall
Email: 
hkeshava@umn.edu

Ph.D. Statistics, University of Michigan

Projects

  • Working on online detection and estimation in high-dimensional regime-switching models with applications in systemic risk analysis and fraud detection.
  • IMA: Scalable detection of abrupt changes in high dimensional, sparse graphical models.
  • Cargill: Using unsupervised machine learning techniques for trading Soybean futures in Chicago Mercantile Exchange (CME).

Related Publications and Working Papers

  1. H. Keshavarz, G. Michailidis, and Y. Atchade, “Sequential change-point detection in high-dimensional Gaussian graphical models”, Submitted to Journal of Machine Learning Research, June 2018.
  2. H. Keshavarz, C. Scott, and X. Nguyen, “Optimal change point detection in Gaussian processes”, Journal of Statistical Planning and Inference, vol. 193, pp. 151-178, 2018.
  3. H. Keshavarz, X. Nguyen, and C. Scott, “Local inversion free estimation of Gaussian spatial processes.”

Related Talks and Events

  1. Participant. QuantCon 2018. New York City, New York. April 2018.
  2. Participant. Citadel Data Open at Michigan, University of Michigan, Ann Arbor. November 2017.
  3. IMA Data Science Seminar. "Online Change-point Detection in High Dimensional Gaussian Graphical Models." Minneapolis, Minnesota. November 28, 2017
  4. IMA Postdoc Seminar. "Online Change-point Detection in High Dimensional Gaussian Graphical Models." Minneapolis, Minnesota. September 29, 2017
  5. IMA Postdoc Seminar. "Local Inversion-Free covariance estimation for Gaussian Spatial Processes." Minneapolis, Minnesota. September 22, 2017