Ch11: A Multiple Equations Approach to Model Based Forecasting

Chapter Summary:

The main focus of this chapter is that real-time macroeconomic forecasting approach. Real-time forecasting is type of short run forecasting, in which analyst publish their forecast earlier than the actual value. This chapter shed light on four important areas of macroeconomic forecasting:

1. Publication of the macroeconomic variable have a great impact on financial market volatility and direction as well. The publication of the variable either through forecasting or through actual publication, both have a significant different impact upon the behaviour of variable.

2. Based upon the first point, it can be easily understand that publication of the forecast value leads to provide opportunities for the decision maker, and then consequently the behaviour of market also change, before the publication of actual value. An individual forecast approach that is better than consensus, because it will provided more opportunities to make a profit or reduce losses.

3. In short-term forecasting (one month ahead), the model should be clearly specify according to the actual release time of the dependent and independent variables. Because if it is not dealt carefully then the forecast result would be misleading, and leads to loss actually (if the forecast result shows the profit).

4. Traditional forecast evaluation methods, such as R-square, adjusted R-square, RMSE, and so on, are necessary, but author recommend an additional step: directional accuracy.

Using the Silvia and Iqbal forecasting approach, we compared the BVAR’s real-time forecast with the Bloomberg real-time consensus and concluded that the BVAR forecasts are more accurate than those of the Bloomberg consensus, on average, for key macro-economic variables.