About:

I recived my Ph.D in Statistics from Michigan State University in June 2002 under the direction of Hira L. Koul. After graduation, I joined in the Department of Mathematics and Statistics at University of New Hampshire. I was promoted to an associate professor in 2008 and professor in 2015. I have taught courses at UNH, such as, Theory of Statistical Inference, Asymptotics of Statistics, Semi-and Nonparametric Regression, Nonparametric Smoothing, Time Series Analysis, etc.

Research:

My research areas mainly include non- and semiparametric regression with correlated errors (or random fields), non- and semiparametric statistical inference using wavelets with thresholding and regression splines with penalty, non- and stationary time series analysis and asymptotic properties of above estimators. In particular, any unknown curve or image (e.g., density function, regression function, spectral density, trend of nonstationary time series, etc.) can be approximated by a series with finite terms with respect to a basis (e.g., wavelets, splines, Fourier, orthogonal series, etc). Once those coefficients among above series are properly estimated, one obtains a non- or semiparametric adaptive or efficient estimate for the curve or regression function. Its optimal properties (e.g., convergence rates, mean integrated squared error, etc.) could be derived using asymptotic theory of statistics. This method could be applied to many other statistical contexts, such as time series analysis for spectral density estimation and its hypothesis testing, adaptive estimation and consistent testing for non- and semiparametric models, statistical inference for nonstationary time series, etc.