Prediction is of fundamental importance in all the sciences, including economics. Forecast accuracy is of obvious importance to users of forecasts because forecasts are used to guide decisions. Comparisons of forecast accuracy are also of importance to economists more generally, who are interested in discriminating among competing economic hypotheses. This project is unified by concern with the forecasting of economic time series. It develops new tools for building forecasting models, constructing point and interval forecasts, evaluating forecast performance, and uncovering empirical macroeconomic regularities. These tools are used in substantive macroeconomic applications. This research is part of the economics of global change initiative because it contributes to the knowledge base of methods needed to enhance the credibility of future efforts to predict the economic impacts of global environmental change or the effects of policies designed to reduce greenhouse gas emissions. This project on modeling and forecasting economic time series has eight interrelated parts: 1) markov-switching models; 2) exactly- unbiased estimators and associated predictors in autoregressive models; 3) low-frequency dynamics and long-memory models; 4) estimation of misspecified forecasting models; 5) comparing predictive accuracy; 6) assessing the predictability of economic time series; 7) forecasting with cointegrated models; and 8) combining forecasts.
