Economics Seminar - "Improving Forecasts in Real TIme"

Economists are constantly looking for stylized facts. One of the most important stylized facts that economists have tried to establish (or disprove) is that forecasts are rational. The theory of rational expectations depends on it, yet the evidence is mixed. Whether a set of forecasts is found to be rational or not seems to depend on many things, including the sample, the source of data on the expectations being examined, and the empirical technique used to investigate rationality. Early papers in the rational-expectations literature used surveys of expectations, such as the Livingston Survey and the Survey of Professional Forecasters (SPF), to test whether the forecasts made by professional forecasters were consistent with the theory. A number of the tests in the 1970s and 1980s cast doubt on the rationality of the forecasts, with notable results by Su and Su (1975) and Zarnowitz (1985). But later results, such as Croushore (2010), found no bias over a longer sample. Similarly, some papers found inefficiency in the forecasts, such as Ball and Croushore (2003), Rudebusch and Williams (2009), and Coibion and Gorodnichenko (2015). The question is: is the bias or inefficiency they found exploitable in real time? In this paper, I develop methods for how, in principle, to improve upon a forecast that is biased or inefficient. The two practical considerations that make forecast improvement difficult to determine are: (1) Is the variable subject to data revisions? (2) Did the bias or inefficiency arise because of a structural change that forecasters did not anticipate?