An Empirical Comparison of Machine Learning Models for Time Series Forecasting

By Nesreen K. Ahmed, Amir F. Atiya, Neamat El Gayar, Hisham El-Shishiny. 2010

764 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Evaluation & Benchmarks, Signal Processing, Time-Series & Forecasting

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Evaluation & Benchmarks

In Signal Processing

In Time-Series & Forecasting

Citations in the corpus

Time-series forecasting with deep learning: a survey (2020, methodology): “While traditional methods have focused on parametric models informed by domain expertise – such as autoregressive (AR) [6], exponential smoothing [7] or structural time series models [8] – modern machine learning methods provide a means to learn temporal dynamics in a purely data-driven manner [9].”
Financial Time Series Prediction Using Deep Learning (2017, methodology): “Due to the inherent uncertainty and non-analytic structure of ﬁnancial markets [5], the task proved to be challenging, where classical linear statistical methods such as the ARIMA model, and statistical machine learning (ML) models have been widely applied [6, 7].”
Data Poisoning Attacks against Autoregressive Models (2016, methodology): “Using machine learning methods for time series forecasting has also gained growing interest, as such methods often outperform traditional ARIMA models from the statistics literature(Ahmed et al. 2010).”
A Bias and Variance Analysis for Multistep-Ahead Time Series Forecasting (2016, neutral): “Alternatively, they could be machine learning models [4], [5].”
Financial Time Series Prediction Based on Deep Learning (2017, background): “Ahmed et al. [6] compared the prediction capabilities of time series with different machine learning algorithms, such as multi-layer perceptron, K-nearest neighbors, classiﬁcation and regression tree, support vector regression, Gaussian process.”
Toward Automatic Time-Series Forecasting Using Neural Networks (2012, methodology): “[27] used the basic form of GRNN and used 10-fold cross-validation to select the best spread factor out of a set of spread factors ranging from 0.”
A review of research into automation in tourism: Launching the Annals of Tourism Research Curated Collection on Artificial Intelligence and Robotics in Tourism (2020, methodology): “In addition to using machine learning techniques to improve accuracy of prediction and forecasting (Ahmed et al., 2010), techniques such as futuring (i.e., envisioning the futures) and future making (Cornish, 2004; Hajer & Pelzer, 2018; Millett, 2006; Szántó, 2018) should be applied to identify…”
Machine Learning Strategies for Time Series Forecasting (2012, background): “In the last two decades, machine learning models have drawn attention and have established themselves as serious contenders to classical statistical models in the forecasting community [1,43,61].”
Forecasting in social settings: The state of the art (2020, methodology): “In a study using 1045 of the monthly M3 series (the same as those utilised by Ahmed et al. 2010) whose number of observations was larger than 81, Makridakis, Spiliotis and Assimakopoulos (2018b) found, using accepted practices to run the methods, that the most accurate of the ML methods were less…”
Forecast combinations of computational intelligence and linear models for the NN5 time series forecasting competition (2011, methodology): “The reason for choosing these values for the learning rate and momentum is that we found them in another large scale study (Ahmed el al [2]) to be quite effective when applying neural networks to time series forecasting.”