Annual dilated convolution neural network for newbuilding ship prices forecasting
Anticipating newbuilding ship prices is crucial for participants in the dynamic shipping market. Although the researchers from forecasting and shipping have shown that the machine learning models outperform statistical ones, convolution neural networks are not investigated. The convolution neural ne...
Saved in:
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/162068 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-162068 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1620682022-10-03T06:32:57Z Annual dilated convolution neural network for newbuilding ship prices forecasting Gao, Ruobin Liu, Jiahui Bai, Xiwen Yuen, Kum Fai School of Civil and Environmental Engineering Engineering::Maritime studies Convolution Neural Network Shipping Market Anticipating newbuilding ship prices is crucial for participants in the dynamic shipping market. Although the researchers from forecasting and shipping have shown that the machine learning models outperform statistical ones, convolution neural networks are not investigated. The convolution neural networks are proposed for image processing, rendering difficulty when handling monthly time series. This paper presents a light neural network with annual dilated convolution filters while extracting the newbuilding market’s short-term and long-term temporal knowledge. The multivariate shipping data are fed into multiple convolutional filters with nonlinear activations. Finally, the convoluted features are fed into a linear layer which maps the features to future values. The annual dilated convolution filter owns a vision across one year and integrates all variables’ temporal information. Besides, the dilation rate renders a parsimonious structure, preventing the model from overfitting. The proposed model is compared with statistical models, Naïve forecasts, and various machine learning models on the newbuilding prices of three tanker markets. The empirical results highlight the superiority of the proposed convolutional neural networks. 2022-10-03T06:32:57Z 2022-10-03T06:32:57Z 2022 Journal Article Gao, R., Liu, J., Bai, X. & Yuen, K. F. (2022). Annual dilated convolution neural network for newbuilding ship prices forecasting. Neural Computing and Applications, 34(14), 11853-11863. https://dx.doi.org/10.1007/s00521-022-07075-x 0941-0643 https://hdl.handle.net/10356/162068 10.1007/s00521-022-07075-x 2-s2.0-85126097728 14 34 11853 11863 en Neural Computing and Applications © 2022 The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Maritime studies Convolution Neural Network Shipping Market |
spellingShingle |
Engineering::Maritime studies Convolution Neural Network Shipping Market Gao, Ruobin Liu, Jiahui Bai, Xiwen Yuen, Kum Fai Annual dilated convolution neural network for newbuilding ship prices forecasting |
description |
Anticipating newbuilding ship prices is crucial for participants in the dynamic shipping market. Although the researchers from forecasting and shipping have shown that the machine learning models outperform statistical ones, convolution neural networks are not investigated. The convolution neural networks are proposed for image processing, rendering difficulty when handling monthly time series. This paper presents a light neural network with annual dilated convolution filters while extracting the newbuilding market’s short-term and long-term temporal knowledge. The multivariate shipping data are fed into multiple convolutional filters with nonlinear activations. Finally, the convoluted features are fed into a linear layer which maps the features to future values. The annual dilated convolution filter owns a vision across one year and integrates all variables’ temporal information. Besides, the dilation rate renders a parsimonious structure, preventing the model from overfitting. The proposed model is compared with statistical models, Naïve forecasts, and various machine learning models on the newbuilding prices of three tanker markets. The empirical results highlight the superiority of the proposed convolutional neural networks. |
author2 |
School of Civil and Environmental Engineering |
author_facet |
School of Civil and Environmental Engineering Gao, Ruobin Liu, Jiahui Bai, Xiwen Yuen, Kum Fai |
format |
Article |
author |
Gao, Ruobin Liu, Jiahui Bai, Xiwen Yuen, Kum Fai |
author_sort |
Gao, Ruobin |
title |
Annual dilated convolution neural network for newbuilding ship prices forecasting |
title_short |
Annual dilated convolution neural network for newbuilding ship prices forecasting |
title_full |
Annual dilated convolution neural network for newbuilding ship prices forecasting |
title_fullStr |
Annual dilated convolution neural network for newbuilding ship prices forecasting |
title_full_unstemmed |
Annual dilated convolution neural network for newbuilding ship prices forecasting |
title_sort |
annual dilated convolution neural network for newbuilding ship prices forecasting |
publishDate |
2022 |
url |
https://hdl.handle.net/10356/162068 |
_version_ |
1746219676439937024 |