arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.08570stat.ME
收藏

跨时间预测协调:最优组合方法与启发式替代方案

Cross-temporal forecast reconciliation: Optimal combination method and heuristic alternatives

Tommaso Di Fonzo, Daniele Girolimetto

AI总结:

本文提出迭代跨时间预测协调程序及最优点预测闭式解,同时处理同期和时间层级约束,并在澳大利亚GDP季度数据上验证其优于单维度协调方法。

摘要:

预测协调是一个后预测过程,旨在提高分层/分组时间序列系统的基础预测质量(Hyndman et al., 2011)。文献中已经考虑了同期(横截面)和时间层级,但除了Kourentzes and Athanasopoulos (2019)之外,这两个特征通常没有被充分同时考虑。本文采用一种能够同时处理两个预测协调维度的记号,展示了两个新结果:(i)一种迭代的跨时间预测协调程序,该程序扩展了Kourentzes and Athanasopoulos (2019)的两步程序,并克服了其一些弱点;(ii)满足同期和时间约束的最优(最小二乘意义下)点预测的闭式表达式。通过对Athanasopoulos et al. (2019)所考虑的澳大利亚GDP收入和支出两侧的95个季度时间序列进行预测实验,研究了所提出程序的可行性,以及与性能最佳的“单维度”(横截面或时间)预测协调程序相比的初步性能评估。

英文摘要:

Forecast reconciliation is a post-forecasting process aimed to improve the quality of the base forecasts for a system of hierarchical/grouped time series (Hyndman et al., 2011). Contemporaneous (cross-sectional) and temporal hierarchies have been considered in the literature, but - except for Kourentzes and Athanasopoulos (2019) - generally these two features have not been fully considered together. Adopting a notation able to simultaneously deal with both forecast reconciliation dimensions, the paper shows two new results: (i) an iterative cross-temporal forecast reconciliation procedure which extends, and overcomes some weaknesses of, the two-step procedure by Kourentzes and Athanasopoulos (2019), and (ii) the closed-form expression of the optimal (in least squares sense) point forecasts which fulfill both contemporaneous and temporal constraints. The feasibility of the proposed procedures, along with first evaluations of their performance as compared to the most performing `single dimension' (either cross-sectional or temporal) forecast reconciliation procedures, is studied through a forecasting experiment on the 95 quarterly time series of the Australian GDP from Income and Expenditure sides considered by Athanasopoulos et al. (2019).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑