6. SIMDAT, SIMEST, and Pseudoreality
Published Online: 29 NOV 2011
Copyright © 2011 John Wiley & Sons, Inc. All rights reserved.
Empirical Model Building: Data, Models, and Reality, Second Edition
How to Cite
Thompson, J. R. (2011) SIMDAT, SIMEST, and Pseudoreality, in Empirical Model Building: Data, Models, and Reality, Second Edition, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9781118109656.ch6
- Published Online: 29 NOV 2011
- Published Print: 24 OCT 2011
Print ISBN: 9780470467039
Online ISBN: 9781118109656
- Dirac comb density estimator;
- stochastic model
The bootstrap is clearly a powerful algorithm for many purposes. However, given the ubiquity of fast computing, it would usually be preferred to use resampling schemes based on better nonparametric density estimators than the Dirac comb. SIMDAT is not a simple resampling so much as it is a stochastic interpolator. We can take the original data and use SIMDAT to generate a SIMDAT pseudodata set of <I>N</I> values. There are now numerous examples in several fields where SIMEST has been used to obtain estimates of the parameters characterizing a marketrelated applied stochastic process. This chapter considers an oncological application to motivate and to explicate SIMEST. It first shows a traditional model-based data analysis, and notes the serious difficulties involved. The chapter then gives a simulation-based, highly computer-intensive way to get what we require to understand the process and act on that understanding.
Controlled Vocabulary Terms