Yuwen Gu

Yuwen Gu

Department of Statistics

University of Connecticut

215 Glenbrook Rd. U-4120

Storrs, CT 06269-4120

Email: yuwen dot gu at uconn dot edu


About Me

I am an assistant professor in the Department of Statistics at the University of Connecticut. I received my PhD degree in statistics from the University of Minnesota and BS degree from the University of Science and Technology of China (USTC).

I work on unconventional regression in high dimensional data analysis. Specifically, I deal with a class of penalized regressions in which the loss functions are not smooth to certain degrees. One particular model I used to study is the sparse asymmetric least squares and my current work is on the sparse quantile and composite quantile regression.

Honors and Awards

  • Bernard Lindgren Graduate Student Teaching Award, University of Minnesota, 2016
  • School of Statistics Alumni Fellowship, University of Minnesota, 2013
  • Guo Moruo Scholarship, USTC, 2010, Summa Cum Laude
  • Outstanding Undergraduate Research Project, USTC, 2010

Publications and Manuscripts

Li, W., Gu, Y. and Liu, L. (2017). Demystifying multiply robust estimators. Under revision.

Gu, Y. and Zou, H. (2017+). Aggregated expectile regression by exponential weighting. To appear in Statistica Sinica.

Gu, Y.*, Jun, F.*, Kong, L., Ma, S. and Zou, H. (2017+). ADMM for high-dimensional sparse penalized quantile regression. To appear in Technometrics.

Gu, Y. and Zou, H. (2016). High-dimensional generalizations of asymmetric least squares regression and their applications. The Annals of Statistics. 44(6), 2661-2694.

Vélez, A., Linehan-Skillings, B. J., Gu, Y., Sun, Y. and Bee, M. (2013). Pulse-number discrimination by Cope’s gray treefrog (Hyla chrysoscelis) in modulated and unmodulated noise. The Journal of the Acoustical Society of America. 134 (4).

*: indicates joint first authorship

Software

  1. R package SALES: Adaptive Lasso and elastic net penalized sparse asymmetric least squares (SALES) and coupled sparse asymmetric least squares (COSALES) regression via coordinate descent and proximal gradient algorithms (available on CRAN)
  2. R package FHDQR: Adaptive lasso and elastic net penalized quantile regression via fast alternating direction method of multipliers algorithms (available here)

Teaching

University of Connecticut

  • STAT/BIST 6494-05 Statistical Learning and Optimization

University of Minnesota

A list of courses I have taught as the primary instructor:

  • STAT 4101 Theory of Statistics I
  • STAT 5401 Applied Multivariate Methods
  • STAT 3011 Introduction to Statistical Analysis
  • STAT 5021 Statistical Analysis

A list of courses I have taught as a teaching assistant:

  • STAT 5601 Nonparametric Methods (Instructor: Dr. Snigdhansu Chatterjee)
  • STAT 3022 Data Analysis
  • STAT 4102 Theory of Statistics II