I am driven by real data problems, and develop theories (if they are relevant), software (if they are useful), and new methods (only when necessary).

Applications in cybersecurity and genetics have led me to study high-dimensional statistics, large-scale multiple testing, extremes values, and algorithms on data streams.

Papers and preprints

• Fundamental Limits of Exact Support Recovery in High Dimensions
  ZG, Stilian Stoev.
  Bernoulli (in press) [accepted version][arXiv]
  - Best Student Paper at [Extreme Value Analysis (EVA) 2019].

• U-PASS: Unified Power Analysis and Forensics for Qualitative Traits in Genetic Association Studies
  ZG, Jonathan Terhorst, Cristopher Van Hout, Stilian Stoev .
  Bioinformatics (in press) [interactive software application] [advance access]
  A related set of slides [on ancillarity, conditioning, and association tests].

• Five Shades of Grey, a.k.a. Phase Transitions in High-dimensional Multiple Testing
  ZG [arXiv]
  - 1st place in the poster competition at [Symposium for the 50-year Anniversary of the Department of Statistics at UM].
  - Rejected at Annals of Statistics. [Referee report 1], [2], [3], and [AE comments].

• On the Rate of Concentration of Maxima in Gaussian Arrays
  Rafail Kartsioukas, ZG, Stilian Stoev.
  Extremes (in press). [arXiv]

• Detection and Identification of Sparse Anomalies in High-dimensional Data Streams
  ZG, Stilian Stoev, Michalis Kallitsis.
  In preparation.

• Fast Online Low-Rank Tensor Decomposition
  ZG, Ruofei Zhao.
  In preparation.

Prior to 2015

• The distribution of incubation and relapse times in experimental human infections with the malaria parasite Plasmodium vivax
  Andrew A Lover, Xiahong Zhao, ZG, Richard J. Coker, and Alex R. Cook.
  BMC Infectious Diseases 2014 14:539. [link]