REAL-ESTATE PRICE PREDICTION UNDER ASYMMETRIC LOSS

This paper deals with the problem of how to adjust a predictive mean i n a practical situation of prediction where there is asymmetry in the loss function. A standard linear model is considered for predicting th e price of real estate using a normal-gamma conjugate prior for the pa rameters. The prior of a subject real estate agent is elicited ut, for comparison, a diffuse prior is also considered. Three loss functions are used: asymmetric linear, asymmetric quadratic and LINEX, and the p arameters under each of these postulated forms are elicited. Theoretic al developments for prediction under each loss function in the presenc e of normal errors are presented and useful tables of adjustment facto r values given. Predictions of the dependent price variable for two pr operties with differing characteristics are made under each loss funct ion and the results compared.