This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 682
Type: Contributed
Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Graphics
Abstract - #307119
Title: Model Assessment for Space-Time Point Processes Using Super-Thinning
Author(s): Robert Alan Clements*+ and Alejandro Veen and Rick Schoenberg
Companies: University of California, Los Angeles and IBM T.J. Watson Research Center and University of California, Los Angeles
Address: University of California - Los Angeles, Los Angeles, CA, 90095,
Keywords: residual analysis ; space-time point processes ; goodness-of-fit ; homogeneous Poisson ; super-thinning
Abstract:

Residual analysis methods for space-time point processes often consist of transforming the observed point process into a residual point process that should be homogeneous Poisson if the fitted model for the conditional intensity is correct. Clustering or inhibition in the residual process is indicative of a lack of fit of the model. Current methods may suffer low power, especially if the conditional intensity is volatile. In this paper, we address this issue with the development of the method of super-thinning, a combination of superposing simulated points and thinning out observed points to homogenize the point process. This technique involves a tuning parameter, k, which gives the user control over how much thinning and superposition is performed to homogenize the process. We also suggest certain criteria for optimizing the performance of super-thinning through the correct choice of k.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.