Evaluating Density Forecasts of Inflation: The Survey of Professional Forecasters

Abstract

Abstract Economic decision makers routinely rely on forecasts to assist their decisions. Until recently, most forecasts were provided only in the form of point forecasts, although forecasters sometimes attached measures of uncertainty, such as standard errors or mean absolute errors, to their forecasts. Recently, the trend has been to accompany point forecasts with a more complete description of the uncertainty of the forecasts, such as explicit interval or density forecasts. An interval forecast indicates the likely range of outcomes by specifying the probability that the actual outcome will fall within a stated interval. The probability may be fixed, at say 0.95, and the associated interval may then vary over time, or the interval may be fixed, as a closed or open interval, and the forecast probability presented, as in the statement that “our estimate of the probability that inflation next year will be below 2.5 percent is p.” A density forecast is stated explicitly as a density or probability distribution. This may be presented analytically, as in “we estimate that next year’s inflation rate is normally distributed around an expected value of 2 percent with a standard deviation of 1 percent,’’ or it may be presented numerically, as when a histogram is reported.