Re-examining the rational expectations hypothesis using panel data on multi-period forecasts

Perspectives on Evaluating Macroeconomic Forecasts

“…A newer and more sophisticated methodology has been developed by Davies andLahiri, (1995, 1998), and has been applied to surveys that contain the forecasts of many individuals or organizations, e.g. Using the Davies-Lahiri methodology it is no longer necessary to confine the evaluation of the predictions obtained from a survey to the "consensus" forecast.…”

A framework for decomposing shocks and measuring volatilities derived from multi-dimensional panel data of survey forecasts

“…Following Davies and Lahiri (1999) Figure 2 shows the relationships among the forecasts with respect to known information and shocks. The brackets above the time line labeled λ 10,4 through λ 5,-1 show the times over which shocks can occur which will affect the accuracy of the six forecasts made at the beginning of quarter 7 for various horizons.…”

“…Davies and Lahiri (1999) relax Batchelor and Dua's (1991) restriction and assume a common bias only across horizons. Because it is reasonable to expect that, whatever inherent bias a given forecaster may exhibit, the bias would decline as the horizon declines, in this model, I relax Davies and Lahiri's (1999) restriction and allow for the possibility of a change in forecast bias over both individuals and horizons.…”

“…Davies and Lahiri describe a methodology for analyzing what they term "multi-dimensional" panel data (i.e. panel data with more than two dimensions) and apply the methodology to the BCS data (Davies and Lahiri, 1995) and the SPF data (Davies and Lahiri, 1999). They show that by employing techniques that account for all three of the data sets' dimensions, additional information can be obtained that would not otherwise be available.…”

Internal Consistency of Survey Respondents' Forecasts: Evidence Based on the Survey of Professional Forecasters*

“…2 The probability forecasts (of the event that output will decline) have received relatively little attention. 3 A seminal paper by Zarnowitz and 2 For example, the point forecasts have been analysed by Zarnowitz (1985), Keane and Runkle (1990), Davies and Lahiri (1999), and the probability distributions by Diebold, Tay and Wallis (1999b), Giordani and Söderlind (2003) and Clements (2006).…”