Researchers have long sought to understand the underlying causes of schizophrenia in the hope of finding a means of prevention. Tramer (1929) noted that psychotic patients were more often born in the late winter or spring. In this talk, I will consider a number of methodological and statistical issues in the design and conduct of studies that attempt to better understand the associations between maternal influenza and schizophrenia, as well as describe other research that is underway to address questions raised by the current H1N1 pandemic.
Infectious disease outbreaks are regular occurrences. It is of great interest to policy makers and public health officials to determine how fast a disease is spreading. This understanding assists in determining what control measures might be effective in containing the disease, if containment is possible. In this talk I will discuss metrics for quantifying infectious disease dynamics in the midst of an outbreak. Statistics plays an important role in estimating these metrics and the uncertainty associated with these estimates. I will illustrate this using the current (recent) Influenza A/H1N1 pandemic within the United States as an example.
Systems biology experiments are increasingly being conducted to discover molecular biomarkers that discriminate between two or more phenotypic classes of interest. Such experiments typically involve measurements of thousands of biomolecular features from each subject in the study – however, the vast majority of the measured entities do not exhibit differences in mean intensity levels between the different classes. The goal of these studies is to identify the subset of features or ‘biomarker set’ that is associated with class membership and a corresponding algorithm that can predict class membership with sufficiently high accuracy. In this talk, I will review some of the computational and statistical challenges arising in the presence of high dimensionality (i.e. number of covariates >> number of subjects). I will describe some popular statistical methods for high dimensional data settings and present results from simulation studies comparing the performance of four commonly used classifiers (K-Nearest Neighbors, Prediction Analysis for Microarrays, Random Forests and Support Vector Machines).
Thanks to the Department of Mathematics and Statistics and the Center for Women in Mathematics for support of the series.
Applied Statistics Lecture series (2008-2009)
Other 5 college seminars of interest:
University of Massachusetts Statistics and
Probability Seminar Series
University of Massachusetts Biostatistics and Epidemiology Seminar Series
Organized by Nicholas Horton.
Last updated March 26, 2010