Suchergebnisse

Compares the predictive performance of artificial neural networks to hedonic pricing models, a more traditional valuation tool. The results document similar predictive performance evidenced from both techniques, which contradicts some of the earlier studies which support a position of artificial neural network superiority. Demonstrates that at least 18 per cent of the "normal" property predictions and over 70 per cent of the "outlier" property predictions contained valuation errors greater than 15 per cent of the actual sales price. The combination of these substantial errors and the model‐optimization costs incurred motivate a message of caution before artificial neural networks are adopted by the real estate valuation and/or lending industries.

Reconsiders the double sinking fund problem by looking at each of
the common methods used. Investigates the underlying assumptions and the
residual errors or inconsistencies. Notes that the use of traditional
dual rate valuations results in a mathematical error within the
valuation and an under‐valuation of the interest. Concludes that the
Double Sinking Fund Method must be recommended in preference to
Pannell′s Method.

This paper provides a model for valuing stocks that takes into account the stochastic processes for earnings and interest rates. Our analysis differs from past research of this type in being applicable to stocks that have a positive probability of zero or negative earnings. By avoiding the singularity at the zero point, our earnings‐based pricing model achieves improved pricing performance. The out‐of‐sample pricing performance of the generalized earnings valuation model (GEVM) and the Bakshi and Chen pricing model are compared on four stocks and two indices. The generalized model has smaller pricing errors and greater parameter stability. Furthermore, deviations between market and model prices tend to be mean‐reverting using the GEVM model, suggesting that the model may be able to identify stock market misvaluation.