Summarising the results of simulation studies

Simulation studies are a powerful tool, but their analysis is not always done well; in particular, Monte Carlo standard errors are often not reported. I present a Stata program, - simsum-, which can output a range of summaries, including bias, precision of one method relative to another, percentage difference between model-based and empirical standard error, power and coverage. Monte Carlo standard errors are computed for all these quantities, using exact or approximate formulae.