Statistical and Machine Learning forecasting methods: Concerns and ways forward.
Statistical and Machine Learning forecasting methods: Concerns and ways forward.
复制标题
DOI：
10.1371/journal.pone.0194889
复制
发表时间：
2018
期刊：
影响因子：
3.7
通讯作者：
Assimakopoulos V
中科院分区：
文献类型：
--
作者：
Makridakis S;Spiliotis E;Assimakopoulos V
机器学习（ML）方法已经在学术文献中提出，作为时间序列预测的统计方法的替代品。然而，很少有证据表明它们在准确性和计算要求方面的相对性能。本文的目的是使用M3竞赛中使用的1045个月度时间序列的大子集来评估多个预测范围的性能。在比较流行的ML方法与八种传统统计方法的样本后准确度之后，我们发现，在所使用的准确度指标和所有预测范围内，前者都占主导地位。此外，我们观察到，它们的计算要求比统计方法大得多。本文讨论了结果，解释了为什么ML模型的准确性低于统计模型，并提出了一些可能的方法。在我们的研究中发现的实证结果强调，需要客观和公正的方法来测试预测方法的性能，可以通过大规模和公开的竞争，允许有意义的比较和明确的结论。
Machine Learning (ML) methods have been proposed in the academic literature as alternatives to statistical ones for time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose of this paper is to evaluate such performance across multiple forecasting horizons using a large subset of 1045 monthly time series used in the M3 Competition. After comparing the post-sample accuracy of popular ML methods with that of eight traditional statistical ones, we found that the former are dominated across both accuracy measures used and for all forecasting horizons examined. Moreover, we observed that their computational requirements are considerably greater than those of statistical methods. The paper discusses the results, explains why the accuracy of ML models is below that of statistical ones and proposes some possible ways forward. The empirical results found in our research stress the need for objective and unbiased ways to test the performance of forecasting methods that can be achieved through sizable and open competitions allowing meaningful comparisons and definite conclusions.
登录
查看更多内容
影响因子：
1.2
作者：
Ahmed, Nesreen K.;Atiya, Amir F.;El-Shishiny, Hisham
通讯作者：
El-Shishiny, Hisham
影响因子：
2.7
作者：
COX, DR;STUART, A
通讯作者：
STUART, A
影响因子：
7.9
作者：
Goodwin, P;Lawton, R
通讯作者：
Lawton, R
影响因子：
7.9
作者：
Assimakopoulos, V;Nikolopoulos, K
通讯作者：
Nikolopoulos, K
影响因子：
3.4
作者：
GARDNER, ES
通讯作者：
GARDNER, ES