Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network

Chia Hung Liao; Te Lun Kao; Shyan Ming Yuan

doi:10.1109/ECICE50847.2020.9301956

Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network

Chia Hung Liao, Te Lun Kao, Shyan Ming Yuan^*

^*Corresponding author for this work

Institute of Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Deep learning has been widely used in many research areas recently. One of the common applications is financial forecasting. Convolutional neural network (CNN), a class of deep learning, which is capable of capturing complex features from the images. In this paper, we proposed a method for forecasting Taiwan capitalization weighted stock index by using Xception, a CNN presented by Chollet, F from Google. Furthermore, with the labeling technique we proposed which aims to predict the median of the closing price of future 20, 30, or 40 days. Based on the results produced by each model, we propose a method to find the optimal trading strategy. Then, after the simulation based on trading the common ETFs in Taiwan which includes 0050.TW, 0056.TW, the promising results show that we can make more profits than those traditional trading strategies such as Buy and Hold and some technical indicators.

Original language	English
Title of host publication	2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020
Editors	Teen-Hang Meen
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	326-329
Number of pages	4
ISBN (Electronic)	9781728180601
DOIs	https://doi.org/10.1109/ECICE50847.2020.9301956
State	Published - 23 Oct 2020
Event	2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020 - Yunlin, Taiwan Duration: 23 Oct 2020 → 25 Oct 2020

Publication series

Name	2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020

Conference

Conference	2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020
Country/Territory	Taiwan
City	Yunlin
Period	23/10/20 → 25/10/20

Keywords

Convolution Neural Network
Gramian Angular Field
Stock Index Prediction
Time Series Analysis

Access to Document

10.1109/ECICE50847.2020.9301956

Cite this

Liao, C. H., Kao, T. L., & Yuan, S. M. (2020). Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network. In T.-H. Meen (Ed.), 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020 (pp. 326-329). Article 9301956 (2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECICE50847.2020.9301956

Liao, Chia Hung ; Kao, Te Lun ; Yuan, Shyan Ming. / Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network. 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020. editor / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 326-329 (2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020).

@inproceedings{ba87ec7b29ae424ca3ee40ab4cfbb864,

title = "Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network",

abstract = "Deep learning has been widely used in many research areas recently. One of the common applications is financial forecasting. Convolutional neural network (CNN), a class of deep learning, which is capable of capturing complex features from the images. In this paper, we proposed a method for forecasting Taiwan capitalization weighted stock index by using Xception, a CNN presented by Chollet, F from Google. Furthermore, with the labeling technique we proposed which aims to predict the median of the closing price of future 20, 30, or 40 days. Based on the results produced by each model, we propose a method to find the optimal trading strategy. Then, after the simulation based on trading the common ETFs in Taiwan which includes 0050.TW, 0056.TW, the promising results show that we can make more profits than those traditional trading strategies such as Buy and Hold and some technical indicators. ",

keywords = "Convolution Neural Network, Gramian Angular Field, Stock Index Prediction, Time Series Analysis",

author = "Liao, {Chia Hung} and Kao, {Te Lun} and Yuan, {Shyan Ming}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; 2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020 ; Conference date: 23-10-2020 Through 25-10-2020",

year = "2020",

month = oct,

day = "23",

doi = "10.1109/ECICE50847.2020.9301956",

language = "English",

series = "2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "326--329",

editor = "Teen-Hang Meen",

booktitle = "2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020",

address = "美國",

}

Liao, CH, Kao, TL & Yuan, SM 2020, Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network. in T-H Meen (ed.), 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020., 9301956, 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020, Institute of Electrical and Electronics Engineers Inc., pp. 326-329, 2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020, Yunlin, Taiwan, 23/10/20. https://doi.org/10.1109/ECICE50847.2020.9301956

Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network. / Liao, Chia Hung; Kao, Te Lun; Yuan, Shyan Ming.
2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020. ed. / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2020. p. 326-329 9301956 (2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network

AU - Liao, Chia Hung

AU - Kao, Te Lun

AU - Yuan, Shyan Ming

PY - 2020/10/23

Y1 - 2020/10/23

N2 - Deep learning has been widely used in many research areas recently. One of the common applications is financial forecasting. Convolutional neural network (CNN), a class of deep learning, which is capable of capturing complex features from the images. In this paper, we proposed a method for forecasting Taiwan capitalization weighted stock index by using Xception, a CNN presented by Chollet, F from Google. Furthermore, with the labeling technique we proposed which aims to predict the median of the closing price of future 20, 30, or 40 days. Based on the results produced by each model, we propose a method to find the optimal trading strategy. Then, after the simulation based on trading the common ETFs in Taiwan which includes 0050.TW, 0056.TW, the promising results show that we can make more profits than those traditional trading strategies such as Buy and Hold and some technical indicators.

AB - Deep learning has been widely used in many research areas recently. One of the common applications is financial forecasting. Convolutional neural network (CNN), a class of deep learning, which is capable of capturing complex features from the images. In this paper, we proposed a method for forecasting Taiwan capitalization weighted stock index by using Xception, a CNN presented by Chollet, F from Google. Furthermore, with the labeling technique we proposed which aims to predict the median of the closing price of future 20, 30, or 40 days. Based on the results produced by each model, we propose a method to find the optimal trading strategy. Then, after the simulation based on trading the common ETFs in Taiwan which includes 0050.TW, 0056.TW, the promising results show that we can make more profits than those traditional trading strategies such as Buy and Hold and some technical indicators.

KW - Convolution Neural Network

KW - Gramian Angular Field

KW - Stock Index Prediction

KW - Time Series Analysis

UR - http://www.scopus.com/inward/record.url?scp=85099558310&partnerID=8YFLogxK

U2 - 10.1109/ECICE50847.2020.9301956

DO - 10.1109/ECICE50847.2020.9301956

M3 - Conference contribution

AN - SCOPUS:85099558310

T3 - 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020

SP - 326

EP - 329

BT - 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020

A2 - Meen, Teen-Hang

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020

Y2 - 23 October 2020 through 25 October 2020

ER -

Liao CH, Kao TL, Yuan SM. Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network. In Meen TH, editor, 2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 326-329. 9301956. (2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020). doi: 10.1109/ECICE50847.2020.9301956

Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this