STATS SLUG

A slug loving Statistics and Data Science

Welcome to STATS SLUG, an academic website created by Cheng-Han Yu, a PhD student at University of California, Santa Cruz (UCSC), who loves Statistics, or its cooler names, machine learning and data science.

Contact information:

About Me

I am a PhD student in Applied Math and Statistics at UC Santa Cruz, advised by Prof. Raquel Prado. Before enjoying California’s sunshine, I studied Economics and Statistics at Indiana University Bloomington (IUB). My current research focuses on Bayesian spatio-temporal modeling and Bayesian variable selection with applications in brain imaging. I am also in the space-time research group of Prof. Hernando Ombao at UC Irvine. My other interests include Bayesian nonparametrics, data visualization and high dimensional problems. Besides analyzing data and coding, I am a foodie, traveling-lover and sports fan, especially MLB, NBA, and professional tennis.

Going down or check my CV to know more about me.

Next

Research

My current project is about detecting activated voxels in a human brain using complex-valued fMRI data. We use spike-and-slab mixtures to develop MCMC and EM algorithms to do variable selections for complex-valued models. Through a simulation study, our proposed methods show that using entire complex-valued data leads to less false positives and more true positives when the noise variance in the underlying model is large than using real, imaginary and magnitude data only.

Work in progress

  1. Bayesian modeling of complex-valued signals and their latent structure, with Prof. Raquel Prado.

Published papers

  1. Feldman, S.R., R. Balkrishnan, H. C. Lin, C. H. Yu, and S.A. Davis (2014). Underuse of Early Follow-Up Visits: A Missed Opportunity to Improve Patients’ Adherence. Journal of Drugs in Dermatology, 13, 833-836.
  2. Ueng, K.L.G. and C. H. Yu (2008). The Neutrality of Profit Tax - Two-Phase Decision Model Under Uncertainty of Economy. Taiwan Public Finance Review, 40, 118-130.
  3. C. H. Yu (2005). Master thesis: The Output and Tax Evasion Decisions of Monopolistic Firm - Under the Conditions of Limited Liability and Market Uncertainty
Next

Teaching

I am currently (winter 2015) a TA for AMS7: Statistical Methods for the Biological, Environmental, and Health Sciences (Prof. Prado). My office hours: Mondays 12pm - 1pm in room Baskin Engineering 312 C/D.

Previous teaching experience

Teaching experience at UCSC

Leading discussion sections
  • Spring 2014: AMS131 Introduction to Probability (Dr. Morris). Introduction to the basic ideas of probability, distribution theory and their applications. Textbook: Probability and Statistics, by M.H. DeGroot and M.J. Schervish. This course covered chapters 1 to 6.
Next

Coursework

I am currently (winter 2015) taking Bayesian Nonparametrics taught by Prof. Rodriguez and Stochastic Differential Equations by Tatiana Xifara.

Previous coursework

Coursework at UCSC

Selected Coursework at IUB

Massive Open Online Courses

Next