Kb. Church et Sp. Curram, FORECASTING CONSUMERS EXPENDITURE - A COMPARISON BETWEEN ECONOMETRIC AND NEURAL-NETWORK MODELS, International journal of forecasting, 12(2), 1996, pp. 255-267
This paper is motivated by the difficulties faced by forecasters in pr
edicting the decline in the growth rate of consumers': expenditure in
the late 1980s. The econometric specifications of four competing expla
nations are replicated and the static forecasts compared with the actu
al outturns. The same data are then used to estimate neural network mo
dels. The main issue is whether the neural network technology can extr
act any more from the data sets provided than the econometric approach
. It is found that the neural network models describe the decline in t
he growth of consumption since the late 1980s as well as, but no bette
r than, the econometric specifications included in the exercise, and a
re shown to be robust when faced with a small number of data points. H
owever, whichever approach is adopted, it is the skill of choosing the
menu of explanatory variables which determines the success of the fin
al results.