Explaining When Deep Learning Models Are Better for Time Series Forecasting

Martín Solís; Luis Alexander Calvo-Valverde

doi:10.3390/engproc2024068001

Explaining When Deep Learning Models Are Better for Time Series Forecasting

Research output: Contribution to journal › Conference article › peer-review

4 Scopus citations

Abstract

There is a gap of knowledge about the conditions that explain why a method has a better forecasting performance than another. Specifically, this research aims to find the factors that can influence deep learning models to work better with time series. We generated linear regression models to analyze if 11 time series characteristics influence the performance of deep learning models versus statistical models and other machine learning models. For the analyses, 2000 time series of M4 competition were selected. The results show findings that can help explain better why a pretrained deep learning model is better than another kind of model.

Translated title of the contribution	Explicación de cuándo los modelos de aprendizaje profundo son mejores para la predicción de series temporales
Original language	English
Article number	1
Pages (from-to)	1-8
Number of pages	8
Journal	Engineering Proceedings
Volume	68
Issue number	1
DOIs	https://doi.org/10.3390/engproc2024068001 https://doi.org/10.3390/engproc2024068001
State	Published - 27 Jun 2024

Keywords

time series
forecast with deep learning
forecast with machine learning

Access to Document

Cite this

@article{cba79d78b6ed4c37878012a73a32784b,

title = "Explaining When Deep Learning Models Are Better for Time Series Forecasting",

abstract = "There is a gap of knowledge about the conditions that explain why a method has a better forecasting performance than another. Specifically, this research aims to find the factors that can influence deep learning models to work better with time series. We generated linear regression models to analyze if 11 time series characteristics influence the performance of deep learning models versus statistical models and other machine learning models. For the analyses, 2000 time series of M4 competition were selected. The results show findings that can help explain better why a pretrained deep learning model is better than another kind of model.",

keywords = "time series, forecast with deep learning, forecast with machine learning",

author = "Mart{\'i}n Sol{\'i}s and Calvo-Valverde, \{Luis Alexander\}",

note = "Publisher Copyright: {\textcopyright} 2024 by the authors.",

year = "2024",

month = jun,

day = "27",

doi = "10.3390/engproc2024068001",

language = "Ingl{\'e}s",

volume = "68",

pages = "1--8",

journal = "Engineering Proceedings",

issn = "2673-4591",

number = "1",

}

TY - JOUR

T1 - Explaining When Deep Learning Models Are Better for Time Series Forecasting

AU - Solís, Martín

AU - Calvo-Valverde, Luis Alexander

PY - 2024/6/27

Y1 - 2024/6/27

N2 - There is a gap of knowledge about the conditions that explain why a method has a better forecasting performance than another. Specifically, this research aims to find the factors that can influence deep learning models to work better with time series. We generated linear regression models to analyze if 11 time series characteristics influence the performance of deep learning models versus statistical models and other machine learning models. For the analyses, 2000 time series of M4 competition were selected. The results show findings that can help explain better why a pretrained deep learning model is better than another kind of model.

AB - There is a gap of knowledge about the conditions that explain why a method has a better forecasting performance than another. Specifically, this research aims to find the factors that can influence deep learning models to work better with time series. We generated linear regression models to analyze if 11 time series characteristics influence the performance of deep learning models versus statistical models and other machine learning models. For the analyses, 2000 time series of M4 competition were selected. The results show findings that can help explain better why a pretrained deep learning model is better than another kind of model.

KW - time series

KW - forecast with deep learning

KW - forecast with machine learning

UR - https://www.scopus.com/pages/publications/85207334298

U2 - 10.3390/engproc2024068001

DO - 10.3390/engproc2024068001

M3 - Artículo de la conferencia

SN - 2673-4591

VL - 68

SP - 1

EP - 8

JO - Engineering Proceedings

JF - Engineering Proceedings

IS - 1

M1 - 1

ER -

Explaining When Deep Learning Models Are Better for Time Series Forecasting

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this