Chapter Summary:
Especially, for the government agent, the long run forecasting is an important element in decision making. Long run forecast is divided in both conditional and unconditional forecast. In case of unconditional long run forecast BVAR model is suggested. Because, this approach used both current and the past values of the data set, and then the output of the forecasts for the target variable is of containing more information.
In case of conditional forecasting approach, a practical option is to build a small-scale macro model that can be utilized to generate forecasts for the target variables.
Both approaches are useful to public and private decision making. The benefit of unconditional forecasting is that it provides neutral forecasts and lets the data speak for itself. Conditional forecasting, in contrast, is an essential tool to analyse the impact of policy changes and/or of an external shock on the economy.
An analyst should become familiar with both approaches and utilize the SAS codes provided in the chapter for long-term forecasting.