Simulation Tools The use of simulation in molecular evolutionary analysis has always been important. As computing speed has increased, the role of simulation-based procedures has become even more prominent. Simulation allows us to test statistical properties of methods, to assess the validity of theoretical asymptotic distributions of statistics, and to study the robustness of procedures to underlying model assumptions. More recently, methods invoking simulation have seen increased use. These techniques include numerical resampling methods for estimating variances or for computing confidence intervals, and also the parametric bootstrap procedures for estimating p-values of test statistics. Hy-Phy provides both parametric and nonparametric simulation tools, and examples of both are illustrated in the following sections.
Sergei L. Kosakovsky Pond and Spencer V. Muse, 1997-2002