How Many Times Should a Stochastic Model Be Run? An Approach Based on Confidence Intervals


A persistent problem in computational cognitive modeling is that many models are stochastic. If a model is stochastic, what is the prediction made by the model? In general, this problem is solved via Monte Carlo simulation. This raises the question of how many runs of the model are adequate to produce a meaningful prediction, a question that has received surprisingly little attention from the community. This paper proposes a systematic approach to the selection of the number of model runs based on confidence intervals and provides tables and computational examples.

Back to Table of Contents