In many forecasting problems, the forecast cost function is used only
in evaluating the forecasts; a second cost function is used in estimat
ing the parameters in the model. In this paper, I explore some of the
ways in which the forecast cost function can be used in estimating the
parameters and, more generally, in producing the forecasts. I define
the optimal forecast and note that it may depend on the entire conditi
onal distribution of the data, which is typically unknown. I then cons
ider three of the steps involved in forming the forecast: approximatin
g the optimal forecast, selecting the model, and estimating any unknow
n parameters. The forecast cost function forms the basis of the approx
imation, selection, and estimation. The methods are illustrated using
time series models applied to 15 US macroeconomic series and in a smal
l Monte Carlo experiment.