The structure of forecasting

The chart on the art of forecasting is from an old Wall Street firm. Some would consider it out of date, but I think it provides a useful framework. While it may be odd to include this in my blog after posting about poor forecasting skills, it is important to understand how forecasting is done. 

A key issue is understanding the difference between precedents and analogies. We can discuss past cases as means of predicting the future. The past provides insight on the future. This is what many economists do with their story-telling. Analogy approach uses models which are based on formal assumptions and propositions. The area of biggest problems for any model forecaster is clarifying his assumptions and determining whether they are correct. As important is understanding the propositions embedded in models. Failure is embedded in poor assumptions or linkages across factors. 

If we are to get better at forecasting, we have to understand the points of failure and whether they can be solved.