A macroeconomist wants to construct forecasts for the following macroeconomic variables: GDP, consumption, investment, government purchases, exports, imports, short-term interest rates, long-term interest rates, and the rate of price inflation. He has quarterly time series for each of these variables from 1970 to 2017. Should he estimate a VAR for these variables and use this for forecasting? Why or why not? Can you suggest an alternative approach?