My core intellectual interest is the way in which parenting behaviors, like the use of physical punishment, or parental expressions of emotional warmth, have an effect on child outcomes like aggression, antisocial behavior, anxiety and depression, and how these dynamics play out across contexts, neighborhoods, and cultures. A lot of my work is done with international samples. In my work I use statistical models, like multilevel models and some econometric models, and software like Stata, R, HLM and ArcGIS, to examine things like growth and change over time, or community, school or parent effects on children and families.
Dr. Liu has a broad research interest in the development of statistical models and techniques to address critical issues in health and nursing sciences, computational processing of Big Data in clinical Informatics and Genomics, statistical modeling and assessment of risk factors (e.g. hypertension, diabetes, central obesity, smoking) for adverse cardiovascular and renal outcomes and maternal and child health. His expertise in statistics includes, but is not limited to, repeated measures models with missing data, multilevel models, latent variable models, and Bayesian and computational statistics. Dr. Liu has led and co-led several NIH-funded projects on the quality of care for hypertensive patients.
I am broadly interested in statistical inference, which is informally defined as the process of turning data into prediction and understanding. I like to work with richly structured data, such as those extracted from texts, images and other spatiotemporal signals. In recent years I have gravitated toward a field in statistics known as Bayesian nonparametrics, which provides a fertile and powerful mathematical framework for the development of many computational and statistical modeling ideas. My motivation for all this came originally from an early interest in machine learning, which continues to be a major source of research interest. A primary focus of my group’s research in machine learning to develop more effective inference algorithms using stochastic, variational and geometric viewpoints.