A University at Buffalo student has been awarded a prestigious award by the Upstate Chapter of the American Statistical Association (ASA).
Dr. Yang Chen, a graduate student in biostatistics at the University at Buffalo School of Public Health, was awarded the Best Young Researchers Award in Bronze for the category of Methodology. Dr. Chen’s research is at the interface of statistics and machine learning and contributes fundamental understanding of the properties and use of statistical distances to address problems in comparative effectiveness research and other areas where direct comparisons are of interest.
For example, in bioinformatics it is of interest to compare microarray data from identical or different tissue types to detect whether there are differences due to the experimental procedures or to detect differences between healthy and diseased tissue.
A second aspect of Dr. Yang Chen’s research pertains the study and development of algorithms for identification of subgroups with differential treatment effect. Responses to a given treatment vary across individuals and treatment effect is not constant. It can differ as a function of patient’s characteristics.
Dr. Chen’s research was conducted under the supervision of Dr. Marianthi Markatou, associate chair of research and healthcare informatics and professor in the department of biostatistics.
“Yang’s work includes methods for subgroup identification of differential treatment effects and highly innovate inferential methods,” explains Dr. Markatou. “This has applications to many problems in biomedicine and public health.”
Statistical distance measures have a large history and play a fundamental role in statistics, machine learning and associated scientific disciplines. However, within the statistical literature, this extensive role has too often been played out behind the scenes, with other aspects of the problems being viewed as more central, more interesting, or more important.
In particular, the prominent role of statistical distances in goodness-of-fit was not recognized until very recently. Dr. Chen’s research develops two-sample goodness-of-fit tests that are useful in big data where the number of observations is massive and the number of variables collected is large. Software for the implementation of the methods constitutes part of the research work where parallel computing techniques facilitate fast computation.
“Ultimately, the goal of my research is to identify predictive markers (covariates) that can explicitly drive changes of treatment effect,” says Dr. Chen. “A summer internship at Pfizer Corporation contributed significantly to my understanding and appreciation of the difficulty of this problem and its importance to pharmaceutical industry in particular.”