Climate and environmental data contribute to the prediction of grain commodity prices using deep learning
Abstract Background Grain commodities are important to people's daily lives and their fluctuations can cause instability for households. Accurate prediction of grain prices can improve food and social security. Methods & Materials This study proposes a hybrid Long Short‐Term Memory (LSTM)‐C...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2023-09-01
|
Series: | Journal of Sustainable Agriculture and Environment |
Subjects: | |
Online Access: | https://doi.org/10.1002/sae2.12041 |
_version_ | 1797690861708902400 |
---|---|
author | Zilin Wang Niamh French Thomas James Calogero Schillaci Faith Chan Meili Feng Aldo Lipani |
author_facet | Zilin Wang Niamh French Thomas James Calogero Schillaci Faith Chan Meili Feng Aldo Lipani |
author_sort | Zilin Wang |
collection | DOAJ |
description | Abstract Background Grain commodities are important to people's daily lives and their fluctuations can cause instability for households. Accurate prediction of grain prices can improve food and social security. Methods & Materials This study proposes a hybrid Long Short‐Term Memory (LSTM)‐Convolutional Neural Network (CNN) model to forecast weekly oat, corn, soybean and wheat prices in the United States market. The LSTM‐CNN is a multivariate model that uses weather data, macroeconomic data, commodities grain prices and snow factors, including Snow Water Equivalent (SWE), snowfall and snow depth, to make multistep ahead forecasts. Results Of all the features, the snow factor is used for the first time for commodity price forecasting. We used the LSTM‐CNN model to evaluate the 5, 10, 15 and 20 weeks ahead forecasting and this hybrid model had the lowest Mean Squared Error (MSE) at 5, 10 and 15 weeks ahead of prediction. In addition, Shapley values were calculated to analyse the feature contribution of the LSTM‐CNN model when forecasting the testing set. Based on the feature contribution, SWE ranked third, fifth and seventh in feature importance in the 5‐week ahead forecast for corn, oats and wheat, respectively, and 7–8 places higher than total precipitation, indicating the potential use of SWE in grain price forecasting. Conclusion The hybrid multivariate LSTM‐CNN model outperformed other models and the newly involved climate data, SWE, showed the research potential of using snow as an input variable to predict grain prices over a multistep ahead time horizon. |
first_indexed | 2024-03-12T02:05:12Z |
format | Article |
id | doaj.art-6fd8bcde98f44b33b056d007140898c5 |
institution | Directory Open Access Journal |
issn | 2767-035X |
language | English |
last_indexed | 2024-03-12T02:05:12Z |
publishDate | 2023-09-01 |
publisher | Wiley |
record_format | Article |
series | Journal of Sustainable Agriculture and Environment |
spelling | doaj.art-6fd8bcde98f44b33b056d007140898c52023-09-07T05:24:24ZengWileyJournal of Sustainable Agriculture and Environment2767-035X2023-09-012325126510.1002/sae2.12041Climate and environmental data contribute to the prediction of grain commodity prices using deep learningZilin Wang0Niamh French1Thomas James2Calogero Schillaci3Faith Chan4Meili Feng5Aldo Lipani6University College London (UCL) London UKWegaw SA Trélex SwitzerlandWegaw SA Trélex SwitzerlandEuropean Commission Joint Research Centre (JRC) Ispra VA ItalySchool of Geographical Sciences University of Nottingham Ningbo China Ningbo ChinaSchool of Geographical Sciences University of Nottingham Ningbo China Ningbo ChinaUniversity College London (UCL) London UKAbstract Background Grain commodities are important to people's daily lives and their fluctuations can cause instability for households. Accurate prediction of grain prices can improve food and social security. Methods & Materials This study proposes a hybrid Long Short‐Term Memory (LSTM)‐Convolutional Neural Network (CNN) model to forecast weekly oat, corn, soybean and wheat prices in the United States market. The LSTM‐CNN is a multivariate model that uses weather data, macroeconomic data, commodities grain prices and snow factors, including Snow Water Equivalent (SWE), snowfall and snow depth, to make multistep ahead forecasts. Results Of all the features, the snow factor is used for the first time for commodity price forecasting. We used the LSTM‐CNN model to evaluate the 5, 10, 15 and 20 weeks ahead forecasting and this hybrid model had the lowest Mean Squared Error (MSE) at 5, 10 and 15 weeks ahead of prediction. In addition, Shapley values were calculated to analyse the feature contribution of the LSTM‐CNN model when forecasting the testing set. Based on the feature contribution, SWE ranked third, fifth and seventh in feature importance in the 5‐week ahead forecast for corn, oats and wheat, respectively, and 7–8 places higher than total precipitation, indicating the potential use of SWE in grain price forecasting. Conclusion The hybrid multivariate LSTM‐CNN model outperformed other models and the newly involved climate data, SWE, showed the research potential of using snow as an input variable to predict grain prices over a multistep ahead time horizon.https://doi.org/10.1002/sae2.12041commodity pricesLSTM‐CNNmultistep ahead predictionsnow water equivalent |
spellingShingle | Zilin Wang Niamh French Thomas James Calogero Schillaci Faith Chan Meili Feng Aldo Lipani Climate and environmental data contribute to the prediction of grain commodity prices using deep learning Journal of Sustainable Agriculture and Environment commodity prices LSTM‐CNN multistep ahead prediction snow water equivalent |
title | Climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
title_full | Climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
title_fullStr | Climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
title_full_unstemmed | Climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
title_short | Climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
title_sort | climate and environmental data contribute to the prediction of grain commodity prices using deep learning |
topic | commodity prices LSTM‐CNN multistep ahead prediction snow water equivalent |
url | https://doi.org/10.1002/sae2.12041 |
work_keys_str_mv | AT zilinwang climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT niamhfrench climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT thomasjames climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT calogeroschillaci climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT faithchan climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT meilifeng climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning AT aldolipani climateandenvironmentaldatacontributetothepredictionofgraincommoditypricesusingdeeplearning |