Bio for Kevin R. Gehringer


In June of 2001 Kevin graduated from the University of Washington with a Ph.D. in Forest Biometrics. His Ph.D. research included predictive dynamic modeling of forest stand development, nonparametric multivariate classification of forest stands, and the simulation of natural and managed forests. In June of 2002 Kevin started a biometrics and statistics consulting business (sole proprietorship) to consult with several clients on the problem of defining multivariate assessment procedures that are both statistically and biologically consistent for their forest management applications. Two years later this work was still going on and had expanded to include the simulation modeling of stream side forests and their ability to produce large woody debris and shade in natural and managed riparian forests, leading Kevin to form Biometrics Northwest LLC, a biometrical and statistical consulting company in September of 2004 (closing his sole proprietorship).


Kevin has been performing biometrical and statistical data analyses, including his Ph.D. research since 1996, and he has been designing, developing, and troubleshooting software since 1987. His professional software development experience includes software development and data analysis with MATLAB, the design, development, and customer support for a multi-platform space science information system developed and maintained by the Jet Propulsion Laboratory, troubleshooting the numerical optimization algorithms used in a retail market analysis and optimization system for MPSI Americas, the development of a graphical user interface for well log data processing for Amoco Production and Research Company, and the development of a prototype client and server for the delivery of geological data over the internet for IBM Systems Integration Division. Kevin has experience programming in MATLAB, R for statistical data analysis, C, Fortran 77, 90, 95, and assembly language on a variety of systems.


Kevinís educational background is in biometrics, statistics, applied mathematics, and numerical computing, and he has performed research in parametric and nonparametric statistical methods, including high dimensional clustering and classification algorithms, probability density estimation, simulation modeling, Monte Carlo analysis, data mining, and numerical analysis and numerical computation including linear and nonlinear differential equations and systems of equations, optimization and parameter estimation. Kevin is interested in data analysis and modeling problems involving large sample sizes, as well as the scalability of clustering and classification algorithms, robust parameter estimation and optimization, predictive modeling, and large scale and distributed computing.


In addition to his Ph.D., Kevin also has an M.S. degree in Applied Mathematics with an emphasis in numerical analysis and numerical computing and a B.S. degree in Mathematics with a minor in Computer Science, both from the University of Tulsa in Tulsa Oklahoma.