People

Thomas O'Gorman

Thomas O'Gorman 

Research Web Page

Dr. O'Gorman is an associate professor who does research in computational statistics and biostatistics.  He earned a Ph.D. in 1988 from the University of Iowa.

After doing some research in the area of adaptive tests of significance, he published several papers in 1996 and 1997 on rank-based adaptive tests.  The problem with rank-based procedures is that they are limited to those relatively simple experimental designs where it is reasonable to assign ranks to the observations.  More complex experimental designs and multiple regressio problems seem to require a different approach.  To address this difficulty, Prof. O'Gorman began researching adaptive methods that used a weighted least squares approach and, in 2004,  he published a book (Applied Adaptive Statistical Methods: Tests of Significance and Confidence Intervals) that described many of the adaptive methods that he had developed at that time.

In 2005, Prof. O'Gorman felt that the permutation approach he had previously used could be improved and developed an adaptive method that used the permutation of residuals approach.  Around that time he also expanded the adaptive methods to include models that had more than one dependent variable.  Recently, he has been working on a comprehensive approach to adaptive teting that would use a single adaptive weighting scheme that can use univariate and multivariate regression models.  Some progress has been made in that direction, but more work needs to be done to verify that these methods are superior to previous methods.

Here are some of Prof. O'Gorman's recent publications:

O'Gorman, T.W.: Applied Adaptive Statistical Methods: Tests of Significance and Confidence Intervals. Society for Industrial and Applied Mathematics, Philadelphia, 2004.

O'Gorman, T.W.: Using Adaptive Methods to Select Variables in Case-Control Studies, Biometrical Journal, 46, 595-605,2004.

O'Gorman, T.W.: The Performance of Randomization Tests that use Permutations of Independent Variables,  Communication in Statistics-Simulation and Computation, 34, 895-908, 2005.

O'Gorman, T. W.: An Adaptive Test for a Subset of Coefficients in a Multivariate regression model. Biometrical Journal, 849-859, 2006.

O'Gorman, T. W.:  An Adaptive Test of a Subset of Regression Coefficients Using Permutations of Residuals, Journal of Statistical Computation and Simulation, 76, 1095-1105, 2006.

O'Gorman, T. W.:  Using Adaptive Tests for the Analysis of Repeated Measurements, Journal of Biopharmaceutical Statistics, 18, 595-610-2008.

O'Gorman, T. W.: An Adaptive Method of Variable Selection in Regression, Communications in Statistics-Simulation and Computation, 37, 1129-1142, 2008.

Prof. O'Gorman's hobbies are biking, hiking and photography.