CU Denver faculty and students at JSM

Data in demand

July 31, 2019

Statisticians are in heavy demand these days. Someone has to mine the data created by biotech, artificial intelligence, sports, clinical trials and just about every other business that compiles data. According to the Bureau of Labor Statistics, employment of mathematicians and statisticians is projected to grow 33% from 2016 to 2026, making it one of the most in-demand jobs on the market.

This week, North America’s largest gathering of statisticians and data scientists gathered at the Joint Statistical Meeting in Denver, where several students and faculty from the University of Colorado Denver presented their innovative research alongside some of the industry’s most brilliant minds.

“In terms of application, the industry is seeing a lot of interest in artificial intelligence and deep learning,” says Audrey Hendricks, PhD, assistant professor of statistics at CU Denver. “Researchers are trying to improve current models and make them better at identifying and correcting for systematic social bias. Ten years ago, data meant thousands of observations. Now it’s in the tens of millions. Data is everywhere, and whether or not you’re trained in the deep theory of it, you’re going to work with it in your career.”

Among their presentations, CU Denver students and faculty introduced research investigating:

  • Racial biases created by the reinforcement-learning AI used by police departments to determine where to send officers at 2 a.m.
  • The fastest statistical method for detecting a disease outbreak in Colorado and elsewhere in the U.S.
  • Hidden ancestries in public genome sites that could lead to misdiagnoses and incorrect genetic associations.

Here’s a full list of CU Denver presentations:

A Comparison of Spatial Scan Methods for Cluster Detection
Mohammad Meysami, University of Colorado Denver; Joshua French, University of Colorado Denver; Lauren M Hall, University of Colorado Denver; Minh Chau Nguyen, University of Colorado Denver; Lee Panter, University of Colorado Denver; Nicholas Weaver, University of Colorado Denver

Analysis of Longitudinal Metabolite Data with Substantial Missingness and Batch Effects
Evan Sticca, University of Colorado Anschutz Medical Campus; Audrey E Hendricks, University of Colorado Denver; Stephanie P Gilley, University of Colorado Anschutz Medical Campus; K Michael Hambidge, University of Colorado Anschutz Medical Campus; Nancy F Krebs, University of Colorado Anschutz Medical Campus; Sarah J Borengasser, University of Colorado Anschutz Medical Campus

Reinforcement Learning as a Solution to Systematic Social Bias in Deep Learning
Kathleen Gatliffe, University of Colorado Denver; Audrey E Hendricks, University of Colorado Denver

Successful and Sustainable Undergraduate Research in Statistics Through Vertical Integration of Experience and Horizontal Integration of Disciplines
Audrey E Hendricks, University of Colorado Denver

Undergraduate Statistics Research: a Viewpoint from a Non-Statistician
Ryan Scherenberg; Megan Sorenson, University of Colorado Denver; Audrey E Hendricks, University of Colorado Denver

Efficient Estimation of Ancestry Proportions Using Genotype Frequencies
Jordan Hall, University of Colorado Denver; Megan Sorenson, University of Colorado Denver; Ryan Scherenberg; Alexandria Ronco, University of Colorado Denver; Yinfei Wu, University of Colorado Denver; James Vance, University of Colorado Denver; Jinyan Lyu, University of Colorado Denver; Christopher Gignoux, University of Colorado Denver; Audrey E Hendricks, University of Colorado Denver

Trans-Ethnic Meta-Analysis of Metabolic Syndrome in a Multi-Ethnic Study Emileigh L. Willems, University of Colorado Denver; Jia Y. Wan, University of California Irvine; Trina M. Norden-Krichmar, University of California Irvine; Karen L. Edwards, University of California Irvine; Stephanie A. Santorico, University of Colorado Denver

Rare Variant Association Tests for Multiple Ancestries Using Common Controls Megan Sorenson, University of Colorado Denver; Audrey E Hendricks, University of Colorado Denver

A Comparison of the Power of Generalized Linear Regressions (GLM) and Generalized Estimating Equations (GEE) in the Phenome –Wide Association Study (PheWAS) Setting
Minh Chau Nguyen, University of Colorado Denver; Erin Austin, University of Colorado Denver

Using Constrained Clustering to Partition Functional MRI Signals Spatiotemporally to Recognize Brain Pattern and BOLD Signals
Aixin Zhang, University of Colorado Denver; Erin Austin, University of Colorado Denver

Multi-Level Monte Carlo Using Quasi-Random Numbers
Lu Vy, University of Colorado Denver; Erin Austin, University of Colorado Denver; Yaning Liu, University of Colorado Denver

Using Push-Forward and Pullback Measures for Parameter Identification and Distribution Estimation
Tian Yu Yen, University of Colorado At Denver; Michael Pilosov, University of Colorado At Denver

A Sandwich Smoother for Spatio-Temporal Arrays and Time Series
Joshua French, University of Colorado Denver; Piotr Kokoszka, Colorado State University