Machine learning advances for time series forecasting

Ricardo Masini, Marcelo Cunha Medeiros, Eduardo F. Mendes · 2021

Read Machine learning advances for time series forecasting by Ricardo Masini, Marcelo Cunha Medeiros, Eduardo F. Mendes (2021) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/openalex-w3114612967.txt.

Description

Ricardo Masini, Marcelo Medeiros, and Eduardo Mendes surveyed machine learning for time series forecasting in the Journal of Economic Surveys in 2021. Time series, from prices to weather, are data ordered by time, and forecasting them is a classic hard problem. The survey covers the newest supervised learning methods.

On the linear side, the authors stress penalized regressions and ensembles of models. On the nonlinear side, they cover neural networks, both shallow and deep, plus tree methods such as random forests and boosted trees. Hybrid models that blend the two families also feature. They review tests for judging which forecast is truly better. Applications in economics and finance close the survey, with an example using high-frequency trading data.

Catalog record

CREATORRicardo Masini, Marcelo Cunha Medeiros, Eduardo F. Mendes
YEAR2021
COLLECTIONPAPERS
FORMATTXT
FILE SIZE104 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEOpenAlex / Journal of Economic Surveys

Source: https://doi.org/10.1111/joes.12429

Browse the full library · About the Tower of Babel