Abstract

VELASQUEZ H., Juan David  and  FRANCO C., Carlos Jaime. Note on predicting consumer price indexes using artificial neuronal networks. Cuad. Adm. [online]. 2010, vol.23, n.41, pp.285-302. ISSN 0120-3592.

This article forecasts the percentage variation in the Colombian Consumer Price Index, using an artificial neuronal network. The model obtained, a multiple-layer perceptron neuronal network, is capable of capturing the seasonal cycle present in the data, by solely using past data as entries into the neuronal network; it is not necessary to include fictitious variables to represent said structural component. This approach is more precise than other models presented in the literature on the topic, including those with exponential smoothing, SARIMA model approximations, and other neuronal network configurations.

Keywords : Prediction; non-linear models; macroeconomics; SARIMA; exponential smoothing.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )