Book Chapters
"Smooth Robust Multi-Horizon Forecasts" with Jennifer L. Castle and David F. Hendry, in A. Chudik, C. Hsiao and A. Timmermann (eds.), Essays in Honor of M. Hashem Pesaran: Prediction and Macro Modeling, Advances in Econometrics (2022), Vol. 34A, Chapter 7, pp. 143-165.
Supplemental Material: Nuffield College Economics Discussion Paper 2021-W01, H.O. Stekler Research Program on Forecasting Working Paper 2020-009
Abstract: We investigate whether smooth robust methods for forecasting can help mitigate pronounced and persistent failure across multiple forecast horizons. We demonstrate that naive predictors are interpretable as local estimators of the long-run relationship with the advantage of adapting quickly after a break, but at a cost of additional forecast error variance. Smoothing over naive estimates helps retain these advantages while reducing the costs, especially for longer forecast horizons. We derive the performance of these predictors after a location shift, and confirm the results using simulations. We apply smooth methods to forecasts of UK productivity and US 10-year Treasury yields and show that they can dramatically reduce persistent forecast failure exhibited by forecasts from macroeconomic models and professional forecasters.
"Evaluating Government Budget Forecasts" with Neil R. Ericsson, in D. Williams, T. Calabrese (eds.), The Palgrave Handbook of Government Budget Forecasting (2019), Chapter 3, pp. 37-69, Palgrave Studies in Public Debt, Spending, and Revenue.
Abstract: This chapter reviews the literature on the evaluation of government budget forecasts, outlines a generic framework for forecast evaluation, and illustrates forecast evaluation with empirical analyses of different U.S. government agencies’ forecasts of U.S. federal debt. Techniques for forecast evaluation include comparison of mean squared forecast errors, forecast encompassing, tests of predictive failure, and tests of bias and efficiency. Recent extensions of these techniques utilize machine-learning algorithms to handle more potential regressors than observations, a characteristic common to big data. These techniques are generally applicable, including to forecasts of components of the government budget; to forecasts of budgets from municipal, state, provincial, and national governments; and to other economic and non-economic forecasts. Evaluation of forecasts is fundamental to assessing the forecasts’ usefulness, and evaluation can indicate ways in which the forecasts may be improved.