On the number of bins in a rank histogram
share
Rank histograms have become popular tools for assessing the reliability of forecasts in meteorology. A forecast that is well-calibrated leads to a uniform rank histogram, and deviations from uniformity indicate miscalibration. However, the ability to identify miscalibration crucially depends on the number of bins chosen for the histogram. If too few bins are chosen, the rank histogram is likely to not detect miscalibrations, if too many are chosen, even perfectly calibrated forecasts lead to rank histograms that do not look uniform. In this paper we address this trade-off and discuss how many bins should be chosen in a rank histogram. Our results indicate that it is often appropriate to choose fewer bins than the usual choice of ensemble size plus one, especially when the number of observations available for verification is small.
READ FULL TEXT