Wenzel, Thomas2004-12-062004-12-061998http://hdl.handle.net/2003/487810.17877/DE290R-5412We consider Pitman-closeness to evaluate the performance of forecasting methods. Optimal weights for the combination of forecasts are calculated with respect to this criterion. We show that theses weights depend on the assuption on the distribution of the forecast errors. In the normal case they are identical with the optimal weights with respect to MSE-criterionenUniversitätsbibliothek Dortmundcombination of forecastsevalutation of forecasting methodsPitman-closenes310Pitman-closeness as a measure to evaluate the equality of forecastsreport