A point forecast (or a prediction) is the expected value of the variable of interest, or the dependent variable, conditional on the given values of the exogenous and predetermined variables. An interval forecast is the confidence interval for the point forecast. In a dynamic model with lagged dependent variable one needs to distinguish between a dynamic and a static forecast. The dynamic forecast for several periods ahead is computed using the previously forecasted values of the dependent variable in every step, whereas the static forecast for several periods ahead amounts to a series of one-step ahead forecasts always using the actual values of the lagged dependent variable. Therefore, the static forecast requires that the data on exogenous and lagged dependent variables be observed for every point in the forecast sample.