COMBINING QUALITATIVE FORECASTS USING LOGIT

This paper introduces a computationally-convenient means of combining qualitative forecasts, through use of legit regression applied to trai ning set data, applicable in dichotomous, polychotomous and ordered po lychotomous contexts. It can be employed in the cases of combining pro bability forecasts, combining qualitative forecasts which have no asso ciated probability forecasts, and combining both of these types of for ecasts, a case for which no combining method currently exists. This me thodology offers insights into the suitability of equal-weight averagi ng of probability forecasts, yields an existing method as a special ca se, and facilitates associated hypothesis testing. (C) 1998 Elsevier S cience B.V.