Computational Ecology and Software (Dec 2011)
A fitter use of Monte Carlo simulations in regression models
Abstract
In this article, I focus on the use of Monte Carlo simulations (MCS) within regression models, being this application very frequent in biology, ecology and economy as well. I'm interested in enhancing a typical fault in this application of MCS, i.e. the inner correlations among independent variables are not used when generating random numbers that fit their distributions. By means of an illustrative example, I provide proof that the misuse of MCS in regression models produces misleading results. Furthermore, I also provide a solution for this topic.