Bayesian composite qualitative forecasting: hog prices again

• Published in: American journal of agricultural economics 1998-08, Vol.80 (3), p.543-551
• Main Author: Dorfman, Jeffrey H.
• Format: Article
• Language: English
• Subjects: Agricultural economics; Bayesian analysis; Bayesian method; C530; CERDO; Commodity prices; composite forecasting; Contingency tables; Economic forecasting models; Economic forecasts; Economic models; EQUATIONS; ETATS UNIS; EUA; FORECASTING; Forecasting models; Forecasts; GDP; GNP; Gross Domestic Product; Gross National Product; hog prices; Hogs; Interest rates; LOGIT MODELS; MATEMATICAS; MATHEMATICS; MATHEMATIQUE; METHODE STATISTIQUE; Methods; METODOS ESTADISTICOS; model uncertainty; MODELE; Modeling; MODELOS; Parametric models; Pigs; Point forecasts; PORCIN; PRECIO AL PRODUCTOR; Prices; PRIX A LA PRODUCTION; PROBABILISTIC MODELS; Probabilities; PROBABILITY; Probability forecasts; PRODUCER PRICES; Q110; Qualitative analysis; qualitative forecasting; Statistical forecasts; STATISTICAL METHODS; SWINE; TECHNIQUE DE PREVISION; TECNICAS DE PREDICCION; Time series; Time series forecasting; UNCERTAINTY; USA; Variables
• ISSN: 0002-9092; 1467-8276
• DOI: 10.2307/1244556

A new method for forming composite qualitative forecasts is presented. A set of qualitative forecasts is evaluated using auxiliary logit models to predict the probability of each forecast's correctness. Individual model forecasts are then combined on the basis of normalized values of these probabilities. This method is demonstrated with three sets of forecasts on the direction of change in hog prices (up or down). The application shows that without any information on the manner in which the individual forecasts are generated this method can form a composite forecast that is superior according to a variety of metrics for evaluating qualitative forecasts.