M. Kaut
Sometimes, the best information about an uncertain future is a single forecast. On the other hand, stochastic-programming models need future data in a form of scenario trees. However, none of the standard scenario-generation methods can generate scenario trees based on a single prediction.
We present a new scenario-generation method that handles the situation, provided one has access to historical forecast errors. The method can take into account dependencies between errors of forecasts of different lengths.
Keywords: stochastic programming, scenario generation
Scheduled
TD1 Finance
May 31, 2016 3:00 PM
Salón de actos
