Loading Events
  • This event has passed.

Statistical Methods for Image Data to Facilitate Geographical Applications

November 17, 2014
3:00 pm - 4:00 pm

Event Navigation

 Dr, Mikyoung Jun
Department of Statistics, Texas A&M University

Associate Professor


In this talk, Dr. Mikyoung Jun will describe a few statistical methodologies for quantifying the structure of variation of images to facilitate geographic applications. In particular, a couple of parametric covariance functions that can be used to model the spatial and spatio-temporal dependence structure in images and statistical estimation methods for their parameters. One of the covariance functions to be discussed, the matern covariance función, has a particular parameter that is linked with fractal index of the surface map. Various statistical methods associated with it, such as having the parameter vary over the image and introducing geometric anisotropy, will be discussed. Some statistical computational techniques to deal with large images will be introduced.


Dr. Mikyoung Jun is an Associate Professor of Statistics at Texas A&M University. She joined TAMU as an Assistant Professor of statistics in 2005. She received PhD in statistics from University of Chicago in 2005. Dr. Jun’s research interest is in spatio-temporal statistics and its application to environmental problems. In particular, her main research expertise lies in covariance modeling of physical processes observed on a global scale, suitable for satellite data and climate model outputs. She is currently the editor (book review) of Journal of Agricultura, Biological, and Environmental Statistics, associate editor of STAT and Journal of the Korean Statistical Society, and node director of STATMOS (Research network for Statistical Methods for Atmospheric and Oceanic Sciences).


November 17, 2014
3:00 pm - 4:00 pm


Koldus Room 110