|
on Forecasting |
| By: | Iania, Leonardo (Université catholique de Louvain, LIDAM/LFIN, Belgium); Algieri, Bernardina (University of Calabria); Leccadito, Arturo (University of Calabria) |
| Abstract: | In recent years, the international community has been increasing its efforts to reduce the human footprint on air pollution and global warming. Total CO2 emissions are a key component of global emission, and as such, they are closely monitored by national and supranational entities. This study evaluates the performance of a broad set of forecasting models and their combinations to predict energy’s carbon dioxide releases using an in-sample and out-of-sample analysis. The focus is on the US for the period 1973-2021 using quarterly observations. The results show that economic variables, energy and interannual climate variability indicators help forecast short-/medium- term CO2 emissions. In addition, a combination of models sharpens quantile predictions. |
| Keywords: | CO2 Emissions ; Forecasting Models ; Quantile Forecast ; Economic, Energy and Nature- related drivers ; Climate Change ; Drought Severity ; Interannual Variability |
| JEL: | C01 C13 C31 C51 C52 C53 C55 C82 Q43 Q47 Q53 Q59 |
| Date: | 2022–05–24 |
| URL: | http://d.repec.org/n?u=RePEc:ajf:louvlf:2022003&r= |
| By: | Cengiz, Doruk; Tekgüç, Hasan |
| Abstract: | We extend the scope of the forecast reconciliation literature and use its tools in the context of causal inference. Researchers are interested in both the average treatment effect on the treated and treatment effect heterogeneity. We show that ex post correction of the counterfactual estimates using the aggregation constraints that stem from the hierarchical or grouped structure of the data is likely to yield more accurate estimates. Building on the geometric interpretation of forecast reconciliation, we provide additional insights into the exact factors determining the size of the accuracy improvement due to the reconciliation. We experiment with U.S. GDP and employment data. We find that the reconciled treatment effect estimates tend to be closer to the truth than the original (base) counterfactual estimates even in cases where the aggregation constraints are non-linear. Consistent with our theoretical expectations, improvement is greater when machine learning methods are used. |
| Keywords: | Forecast Reconciliation; Non-linear Constraints; Causal Machine Learning Methods; Counterfactual Estimation; Difference-in-Differences |
| JEL: | C53 |
| Date: | 2022–06 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:114478&r= |