Suchergebnisse

A decision maker doubts the stationarity of his environment. In response, he uses two models, one with time-varying parameters, and another with constant parameters. Forecasts are then based on a Bayesian model averaging strategy, which mixes forecasts from the two models. In reality, structural parameters are constant, but the (unknown) true model features expectational feedback, which the reduced-form models neglect. This feedback permits fears of parameter instability to become self-confirming. Within the context of a standard asset-pricing model, we use the tools of large deviations theory to show that even though the constant parameter model would converge to the rational expectations equilibrium if considered in isolation, the mere presence of an unstable alternative drives it out of consideration. (JEL C63, D83, D84)

We present ensemble Bayesian model averaging (EBMA) and illustrate its ability to aid scholars in the social sciences to make more accurate forecasts of future events. In essence, EBMA improves prediction by pooling information from multiple forecast models to generate ensemble predictions similar to a weighted average of component forecasts. The weight assigned to each forecast is calibrated via its performance in some validation period. The aim is not to choose some "best" model, but rather to incorporate the insights and knowledge implicit in various forecasting efforts via statistical postprocessing. After presenting the method, we show that EBMA increases the accuracy of out-of-sample forecasts relative to component models in three applied examples: predicting the occurrence of insurgencies around the Pacific Rim, forecasting vote shares in U.S. presidential elections, and predicting the votes of U.S. Supreme Court Justices.

This paper offers two innovations for empirical growth research. First, the paper discusses principal components augmented regressions to take into account all available information in well-behaved regressions. Second, the paper proposes a frequentist model averaging framework as an alternative to Bayesian model averaging approaches. The proposed methodology is applied to three data sets, including the Sala-i-Martin et al. (2004) and Fernandez et al. (2001) data as well as a data set of the European Union member states' regions. Key economic variables are found to be significantly related to economic growth. The findings highlight the relevance of the proposed methodology for empirical economic growth research.