Fabian Krüger

Using Entropic Tilting to Combine BVAR Forecasts with External Nowcasts

This article shows entropic tilting to be a flexible and powerful tool for combining medium-term forecasts from BVARs with short-term forecasts from other sources (nowcasts from either surveys or other models). Tilting systematically improves the accuracy of both point and density forecasts, and tilting the BVAR forecasts based on nowcast means and variances yields slightly greater gains in density accuracy than does just tilting based on the nowcast means. Hence, entropic tilting can offer—more so for persistent variables than not-persistent variables—some benefits for accurately estimating the uncertainty of multi-step forecasts that incorporate nowcast information.

Disagreement versus uncertainty: Evidence from distribution forecasts.

We use a cross-section of economic survey forecasts to predict the distribution of US macro variables in real time. This generalizes the existing literature, which uses disagreement (i.e., the cross-sectional variance of survey forecasts) to predict uncertainty (i.e., the conditional variance of future macroeconomic quantities). Our results show that cross-sectional information can be helpful for distribution forecasting, but this information needs to be modeled in a statistically efficient way in order to avoid overfitting. A simple one-parameter model which exploits time variation in the cross-section of survey point forecasts is found to perform well in practice.This paper generalizes the discussion about disagreement versus uncertainty in macroeconomic survey data by emphasizing the importance of the (unknown) true predictive density. Using a forecast combination approach, we ask whether cross sections of survey point forecasts help to approximate the true predictive density. We find that although these cross-sections perform poorly individually, their inclusion into combined predictive densities can significantly improve upon densities relying solely on time series information.

Forecasting Conditional Probabilities of Binary Outcomes under Misspecification

We consider constructing probability forecasts from a parametric binary choice model under a large family of loss functions (“scoring rules”). Scoring rules are weighted averages over the utilities that heterogeneous decision makers derive from a publicly announced forecast (Schervish, 1989). Using analytical and numerical examples, we illustrate howdifferent scoring rules yield asymptotically identical results if the model is correctly specified. Under misspecification, the choice of scoring rule may be inconsequential under restrictive symmetry conditions on the data-generating process. If these conditions are violated, typically the choice of a scoring rule favors some decision makers over others.