Reproducible Econometric Simulations

Reproducibility of economic research has attracted considerable attention in recentyears. So far, the discussion has focused on reproducibility of empirical analyses. Thispaper addresses a further aspect of reproducibility, the reproducibility of computationalexperiments. We examine the current situation in econometrics and derive a set of guidelinesfrom our ndings. To illustrate how computational experiments could be conductedand reported we present an example from time series econometrics that explores thenite-sample power of certain structural change tests.