Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports

Deep neural networks (DNNs) are prominent in predictive analytics for accurately forecasting target variables. However, inherent uncertainties necessitate constructing prediction intervals for reliability. The existing literature often lacks practical methodologies for creating predictive intervals,...

Full description

Saved in:
Bibliographic Details
Main Authors: Unyamanee Kummaraka, Patchanok Srisuradetchai
Format: Article
Language:English
Published: MDPI AG 2024-08-01
Series:Forecasting
Subjects:
Online Access:https://www.mdpi.com/2571-9394/6/3/33
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850261329801117696
author Unyamanee Kummaraka
Patchanok Srisuradetchai
author_facet Unyamanee Kummaraka
Patchanok Srisuradetchai
author_sort Unyamanee Kummaraka
collection DOAJ
description Deep neural networks (DNNs) are prominent in predictive analytics for accurately forecasting target variables. However, inherent uncertainties necessitate constructing prediction intervals for reliability. The existing literature often lacks practical methodologies for creating predictive intervals, especially for time series with trends and seasonal patterns. This paper explicitly details a practical approach integrating dual-output Monte Carlo Dropout (MCDO) with DNNs to approximate predictive means and variances within a Bayesian framework, enabling forecast interval construction. The dual-output architecture employs a custom loss function, combining mean squared error with Softplus-derived predictive variance, ensuring non-negative variance values. Hyperparameter optimization is performed through a grid search exploring activation functions, dropout rates, epochs, and batch sizes. Empirical distributions of predictive means and variances from the MCDO demonstrate the results of the dual-output MCDO DNNs. The proposed method achieves a significant improvement in forecast accuracy, with an RMSE reduction of about 10% compared to the seasonal autoregressive integrated moving average model. Additionally, the method provides more reliable forecast intervals, as evidenced by a higher coverage proportion and narrower interval widths. A case study on Thailand’s durian export data showcases the method’s utility and applicability to other datasets with trends and/or seasonal components.
format Article
id doaj-art-9be8a7b56596485f91928e854fad4c3b
institution OA Journals
issn 2571-9394
language English
publishDate 2024-08-01
publisher MDPI AG
record_format Article
series Forecasting
spelling doaj-art-9be8a7b56596485f91928e854fad4c3b2025-08-20T01:55:27ZengMDPI AGForecasting2571-93942024-08-016361663610.3390/forecast6030033Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian ExportsUnyamanee Kummaraka0Patchanok Srisuradetchai1Department of Statistics, Faculty of Science, Chiang Mai University, Chiang Mai 50200, ThailandDepartment of Mathematics and Statistics, Faculty of Science and Technology, Thammasat University, Pathum Thani 12120, ThailandDeep neural networks (DNNs) are prominent in predictive analytics for accurately forecasting target variables. However, inherent uncertainties necessitate constructing prediction intervals for reliability. The existing literature often lacks practical methodologies for creating predictive intervals, especially for time series with trends and seasonal patterns. This paper explicitly details a practical approach integrating dual-output Monte Carlo Dropout (MCDO) with DNNs to approximate predictive means and variances within a Bayesian framework, enabling forecast interval construction. The dual-output architecture employs a custom loss function, combining mean squared error with Softplus-derived predictive variance, ensuring non-negative variance values. Hyperparameter optimization is performed through a grid search exploring activation functions, dropout rates, epochs, and batch sizes. Empirical distributions of predictive means and variances from the MCDO demonstrate the results of the dual-output MCDO DNNs. The proposed method achieves a significant improvement in forecast accuracy, with an RMSE reduction of about 10% compared to the seasonal autoregressive integrated moving average model. Additionally, the method provides more reliable forecast intervals, as evidenced by a higher coverage proportion and narrower interval widths. A case study on Thailand’s durian export data showcases the method’s utility and applicability to other datasets with trends and/or seasonal components.https://www.mdpi.com/2571-9394/6/3/33dual-output neural networkMonte Carlo Dropoutforecast intervaltime-series forecastingregularization
spellingShingle Unyamanee Kummaraka
Patchanok Srisuradetchai
Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
Forecasting
dual-output neural network
Monte Carlo Dropout
forecast interval
time-series forecasting
regularization
title Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
title_full Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
title_fullStr Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
title_full_unstemmed Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
title_short Time-Series Interval Forecasting with Dual-Output Monte Carlo Dropout: A Case Study on Durian Exports
title_sort time series interval forecasting with dual output monte carlo dropout a case study on durian exports
topic dual-output neural network
Monte Carlo Dropout
forecast interval
time-series forecasting
regularization
url https://www.mdpi.com/2571-9394/6/3/33
work_keys_str_mv AT unyamaneekummaraka timeseriesintervalforecastingwithdualoutputmontecarlodropoutacasestudyondurianexports
AT patchanoksrisuradetchai timeseriesintervalforecastingwithdualoutputmontecarlodropoutacasestudyondurianexports