An Empirical Comparison of Machine Learning Models for Time Series Forecasting
An Empirical Comparison of Machine Learning Models for Time Series Forecasting
复制标题
DOI：
10.1080/07474938.2010.481556
复制
发表时间：
2010-01-01
影响因子：
1.2
通讯作者：
El-Shishiny, Hisham
中科院分区：
文献类型：
--
作者：
Ahmed, Nesreen K.;Atiya, Amir F.;El-Shishiny, Hisham
在这项工作中，我们提出了一个大规模的时间序列预测的主要机器学习模型的比较研究。具体而言，我们将模型应用于每月的M3时间序列竞争数据（约一千个时间序列）。对于回归或时间序列预测问题的机器学习模型，很少有大规模的比较研究，所以我们希望这项研究能够填补这一空白。考虑的模型是多层感知器，贝叶斯神经网络，径向基函数，广义回归神经网络（也称为核回归），K-最近邻回归，CART回归树，支持向量回归和高斯过程。该研究揭示了不同方法之间的显着差异。最好的两种方法是多层感知器和高斯过程回归。除了模型比较之外，我们还测试了不同的预处理方法，并表明它们对性能有不同的影响。
In this work we present a large scale comparison study for the major machine learning models for time series forecasting. Specifically, we apply the models on the monthly M3 time series competition data (around a thousand time series). There have been very few, if any, large scale comparison studies for machine learning models for the regression or the time series forecasting problems, so we hope this study would fill this gap. The models considered are multilayer perceptron, Bayesian neural networks, radial basis functions, generalized regression neural networks (also called kernel regression), K-nearest neighbor regression, CART regression trees, support vector regression, and Gaussian processes. The study reveals significant differences between the different methods. The best two methods turned out to be the multilayer perceptron and the Gaussian process regression. In addition to model comparisons, we have tested different preprocessing methods and have shown that they have different impacts on the performance.