Mathematics and Statistics Colloquium (Online)

  • November 4, 2021
  • 4:30 PM - 5:30 CST
  • Zoom (online)
  • Tuyen Tran, ttran18@luc.edu
  • Not open to the public.
  • https://luc.zoom.us/j/81081283725
  • Add to calendar
  • Details

    Speaker: Hakan Demirtas (University of Illinois at Chicago)

    Title: Hybrid data generation

    Abstract: This talk formulates a plan for implementing a unified, general-purpose mixed data generation framework that includes nearly all major types of variables (i.e., binary, ordinal, count, and continuous) when the marginal distributions and a feasible association structure in the form of Pearson or Spearman correlations are specified for simulation purposes via a mathematically unsophisticated but effective sorting procedure.

    Most relevant papers:
    Demirtas, H. & Hedeker, D. (2011). A practical way for computing approximate lower and upper correlation bounds. American Statistician, Volume 65, Issue 2, 104-109.
    Demirtas, H. (2019). Inducing any feasible level of correlation to bivariate data with any marginals. American Statistician, Volume 73, Issue 3, 273-277.

    About the speaker: Dr. Demirtas is an Associate Professor of Biostatistics at the Division of Epidemiology and Biostatistics in University of Illinois at Chicago. His research interests include missing data, statistical computing, multiple imputation, stochastic simulation, and random number generation