A new method for forming composite qualitative forecasts is presented.
A set of qualitative forecasts is evaluated using auxiliary legit mod
els to predict the probability of each forecast's correctness. Individ
ual model forecasts are then combined on the basis of normalized value
s of these probabilities. This method is demonstrated with three sets
of forecasts on the direction of change in hog prices (up or down). Th
e application shows that without any information on the manner in whic
h the individual forecasts are generated this method can form a compos
ite forecast that is superior according to a variety of metrics for ev
aluating qualitative forecasts.