Created 23/10/2011 09:45:05 AM

There are some points in our reading units that should be memorized as well as golden verses. I think that one of those verses is J. Scott Armstrong’s golden rules of forecasting (2005) as they help us increase the level of accuracy in our forecasts:

1. Match the forecasting method to the situation

2. Use domain knowledge

3. Structure the problem

4. Model experts’ forecasts

5. Represent the problem realistically

6. Use causal models when you have good information

7. Use simple quantitative methods

8. Be conservative when uncertain

9. Combine forecasts

Regardless of using these rules, we should learn how to choose our forecasting method between two major approaches: judgmental and quantitative based on the amount of objective data we have. However, we may reach a point where it’s better to combine several methods to achieve better forecasts. The competitive benefit with judgmental methods is that they may show experts how they are weighting various factors.

For instance, in Judgmental bootstrapping method we make a statistical inference of a judge’s model by running a regression of the judgmental forecasts against the information that the forecaster used. The reasoning of this method is that the model applies the man’s rules more consistently than the man can. Bootstrapping also allows us to estimate the effects of changing key variables when we have no historical data on such changes and thus avoid assessing them with econometric methods.     

Reference:

Armstrong, J. S. (2005, June). The forecasting canon: Nine generalizations to improve forecast accuracy. Foresight: The International Journal of Applied Forecasting, 1(1), 29-35.