|
on Forecasting |
| By: | Matteo Mogliani (BANQUE DE FRANCE); Florens Odendahl (BANCO DE ESPAÑA AND CEMFI) |
| Abstract: | The common choice of using a direct forecasting scheme implies that the individual predictions ignore information on their cross-horizon dependence. However, this dependence is needed if the forecaster has to construct, based on direct density forecasts, predictive objects that are functions of several horizons (e.g. when constructing annual-average growth rates from quarter-on-quarter growth rates). To address this issue we propose using copulas to combine the individual h-step-ahead predictive distributions into one joint predictive distribution. Our method is particularly appealing to those for whom changing the direct forecasting specification is too costly. We use a Monte Carlo study to demonstrate that our approach leads to a better approximation of the true density than an approach that ignores the potential dependence. We show the superior performance of our method using several empirical examples, where we construct (i) quarterly forecasts using month-on-month direct forecasts, (ii) annual-average forecasts using monthly year-on-year direct forecasts, and (iii) annual-average forecasts using quarter-on-quarter direct forecasts. |
| Keywords: | joint predictive distribution, frequency transformation, path forecasts, cross-horizon dependence |
| JEL: | C53 C32 E37 |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:bde:wpaper:2511 |
| By: | William A. Brock (Department of Economics, University of Wisconsin-Madison); J. Isaac Miller (Department of Economics, University of Missouri) |
| Abstract: | Nearly one half of the positive feedback mechanisms that are identified in the literature as potential tipping elements in the climate system and are of serious concern within the next century are located in the northern part of the Northern Hemisphere. Improving forecasts of northern temperatures is therefore critical to improving our understanding and perhaps early detection of tipping points. We propose forecasting northern temperatures using a structural geophysical model of polar amplification, which is defined as the acceleration of warming in regions closer to the poles and the North Pole in particular, that uses anthropogenic forcing and southern temperatures as covariates. We show using pseudo-out-of-sample forecasts over a range of time periods that this geophysical model improves medium-run forecasts over otherwise similar benchmark forecasting models. Using this model, we forecast temperature anomalies in the northern part of the Northern Hemisphere to increase from 1.861C (over the 1961-1990 baseline) in 2023 to 2.214C with a 95% forecast interval of (1.399, 3.147) C by 2035. |
| Keywords: | climate change, polar amplification, moist energy balance model, statistical forecasting |
| JEL: | C32 C33 C53 Q54 |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:umc:wpaper:2502 |