Qing (Wendy) Wang
Associate Professor of Mathematics
Links
Statistician specializing in topics of U-statistics, nonparametric kernel density estimation, risk estimation, variance estimation, cross-validation, resampling schemes, and extrapolation/interpolation methods.
I was born and raised in Beijing, China. My academic journey in the U.S. began in 2007 when I started the PhD program in statistics at Penn State University. I completed my PhD at Penn State in 2012 under the supervision of the late Dr. Bruce G. Lindsay. Prior to joining Wellesley, I held tenure-track Assistant Professor of Statistics positions at Williams College (2012–2015) and Bentley University (2015–2016).
My research centers on topics in U-statistics, risk and variance estimation, and cross-validation. My current research interests also include penalization methods, model assessment tools, high-dimensional mediation analysis, jackknife empirical likelihood, and multilabel classification.
I teach both applied and theoretical statistics courses, including Introductory Statistics, Regression Analysis, Probability, Multivariate Data Analysis, Nonparametric Statistics, and Bayesian Statistics. As one of the data science advisors, I also rotate to teach the data science senior capstone course. I deeply enjoy mentoring undergraduate students on research projects and have a solid and growing record of publications with undergraduate co-authors. I also have a great passion for undergraduate statistics education, and I’m actively involved in developing and incorporating active-learning educational tools in my classroom teaching.
Outside of my academic life, I enjoy baking, cooking, practicing yoga, traveling with my husband, and spending time with my two lovely children, Iwa and Ian.
Education
- B.S., Beijing Normal University (北京师范大学)
- M.S., Pennsylvania State University-Main Campus
- Ph.D., Pennsylvania State University-Main Campus
Current and upcoming courses
Probability
MATH220
Probability is the mathematics of uncertainty. We begin by developing the basic tools of probability theory, including counting techniques, conditional probability, and Bayes's Theorem. We then survey several of the most common discrete and continuous probability distributions (binomial, Poisson, uniform, normal, and exponential, among others) and discuss mathematical modeling using these distributions. Often we cannot calculate probabilities exactly, and we need to approximate them. A powerful tool here is the Central Limit Theorem, which provides the link between probability and statistics. Another strategy when exact results are unavailable is simulation. We examine Markov chain Monte Carlo methods, which offer a means of simulating from complicated distributions.
(MATH 220 and STAT 220 are cross-listed courses.)-
Regression Analysis and Statistical Models
STAT318
This is an applied regression analysis course that involves hands-on data analysis. Topics covered during the semester include simple and multiple linear regression models, model diagnostics and remedial measures, matrix representation of linear regression models, model comparison and selection, generalized linear regression models (e.g. binary logistic regression, multinomial logistic regression, ordinal logistic regression, and Poisson regression). Statistical language R will be used throughout the course to realize fitting linear (or generalized linear) regressions models, model diagnostics, model comparison and selection, and simulations.