Applied Time Series Analysis for the Social Sciences

11The LSE Approach: Encompassing, General‐to‐Specific Modeling, and Forecasting Success

The applied econometrician is like a farmer who notices that the yield is somewhat higher under trees where birds roost, and he uses this as evidence that bird droppings increase yields. However another farmer objects that he used the same data but came up with the conclusion that moderate amounts of shade increase yields.

(Leamer, 1983: 31)

General‐to‐specific modeling is a hallmark of the LSE School approach to time series analysis, and its introduction in Chapters 8–10 likely seemed a radical departure from the ARIMA modeling covered in Chapters 4–7. In 1978, however, James Davidson, David Hendry, Frank Srba, and Stephen Yeo (DHSY) published a canonical paper that “illustrate[d] the advantage of using a wide range of different techniques (including both ‘econometric’ and ‘time series’ methods) when analyzing aggregate data.”¹ While they examined autocorrelation functions and consciously modeled the autocorrelation of their series in the tradition of ARIMA modelers, they applied the principle of “general‐to‐specific” modeling in the same way that it was applied in the COMFAC case and tested, ...