University of California - Department of Statistics
The talk will survey various statistical techniques for evaluating the goodness-of-fit of point process models, such as those used in earthquake forecasting. Commonly used methods involve examining spatial aggregates over pre-specified grid cells as described in detail by Baddeley, Turner, MÃ¸ller and Hazelton (2005). These methods can have severe drawbacks, such as loss of power when large grid cells are used, and enormous variance in the case of small grid cells. In the case of assessing spatial-temporal models for phenomena such as earthquakes, alternative diagnostics may be preferable, such as rescaled and thinned residuals and weighted second-order statistics. These residual techniques will be described and illustrated here, and their pros and cons for the purpose of assessing models for earthquake forecasting will be discussed.