Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks

Ajla Kulaglic; Burak Berk Ustundag

doi:10.1109/UBMK.2018.8566614

Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks

Ajla Kulaglic, Burak Berk Ustundag

Department of Computer Engineering

Istanbul Technical University

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

8 Citations (Scopus)

Abstract

This paper presents a highly reliable and accurate stock-price prediction model. We aim to anticipate the stock price with respect to multiple patterns in different time scales. The stock price time-series are decomposed, using discrete wavelet transform (DWT), into temporal resolution of varying scales. Then, each subseries is used to predict the stock price using two types of neural network (NN) models with one and two hidden layers. Results show that having multiple time windows in input datasets together with DWT decrease the RMSE of NN models below 10%.

Original language	English
Title of host publication	UBMK 2018 - 3rd International Conference on Computer Science and Engineering
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	518-521
Number of pages	4
ISBN (Electronic)	9781538678930
DOIs	https://doi.org/10.1109/UBMK.2018.8566614
Publication status	Published - 6 Dec 2018
Event	3rd International Conference on Computer Science and Engineering, UBMK 2018 - Sarajevo, Bosnia and Herzegovina Duration: 20 Sept 2018 → 23 Sept 2018

Publication series

Name	UBMK 2018 - 3rd International Conference on Computer Science and Engineering

Conference

Conference	3rd International Conference on Computer Science and Engineering, UBMK 2018
Country/Territory	Bosnia and Herzegovina
City	Sarajevo
Period	20/09/18 → 23/09/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

Apple Stock Prices
Discrete wavelet transform
financial time series data
neural networks
stock price forecasting

Access to Document

10.1109/UBMK.2018.8566614

Cite this

Kulaglic, A., & Ustundag, B. B. (2018). Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks. In UBMK 2018 - 3rd International Conference on Computer Science and Engineering (pp. 518-521). Article 8566614 (UBMK 2018 - 3rd International Conference on Computer Science and Engineering). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/UBMK.2018.8566614

Kulaglic, Ajla ; Ustundag, Burak Berk. / Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks. UBMK 2018 - 3rd International Conference on Computer Science and Engineering. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 518-521 (UBMK 2018 - 3rd International Conference on Computer Science and Engineering).

@inproceedings{35ff3572ad3a4e89875009bd4daf35c8,

title = "Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks",

abstract = "This paper presents a highly reliable and accurate stock-price prediction model. We aim to anticipate the stock price with respect to multiple patterns in different time scales. The stock price time-series are decomposed, using discrete wavelet transform (DWT), into temporal resolution of varying scales. Then, each subseries is used to predict the stock price using two types of neural network (NN) models with one and two hidden layers. Results show that having multiple time windows in input datasets together with DWT decrease the RMSE of NN models below 10%.",

keywords = "Apple Stock Prices, Discrete wavelet transform, financial time series data, neural networks, stock price forecasting",

author = "Ajla Kulaglic and Ustundag, {Burak Berk}",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 3rd International Conference on Computer Science and Engineering, UBMK 2018 ; Conference date: 20-09-2018 Through 23-09-2018",

year = "2018",

month = dec,

day = "6",

doi = "10.1109/UBMK.2018.8566614",

language = "English",

series = "UBMK 2018 - 3rd International Conference on Computer Science and Engineering",

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

pages = "518--521",

booktitle = "UBMK 2018 - 3rd International Conference on Computer Science and Engineering",

address = "United States",

}

Kulaglic, A & Ustundag, BB 2018, Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks. in UBMK 2018 - 3rd International Conference on Computer Science and Engineering., 8566614, UBMK 2018 - 3rd International Conference on Computer Science and Engineering, Institute of Electrical and Electronics Engineers Inc., pp. 518-521, 3rd International Conference on Computer Science and Engineering, UBMK 2018, Sarajevo, Bosnia and Herzegovina, 20/09/18. https://doi.org/10.1109/UBMK.2018.8566614

Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks. / Kulaglic, Ajla; Ustundag, Burak Berk.
UBMK 2018 - 3rd International Conference on Computer Science and Engineering. Institute of Electrical and Electronics Engineers Inc., 2018. p. 518-521 8566614 (UBMK 2018 - 3rd International Conference on Computer Science and Engineering).

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

TY - GEN

T1 - Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks

AU - Kulaglic, Ajla

AU - Ustundag, Burak Berk

PY - 2018/12/6

Y1 - 2018/12/6

N2 - This paper presents a highly reliable and accurate stock-price prediction model. We aim to anticipate the stock price with respect to multiple patterns in different time scales. The stock price time-series are decomposed, using discrete wavelet transform (DWT), into temporal resolution of varying scales. Then, each subseries is used to predict the stock price using two types of neural network (NN) models with one and two hidden layers. Results show that having multiple time windows in input datasets together with DWT decrease the RMSE of NN models below 10%.

AB - This paper presents a highly reliable and accurate stock-price prediction model. We aim to anticipate the stock price with respect to multiple patterns in different time scales. The stock price time-series are decomposed, using discrete wavelet transform (DWT), into temporal resolution of varying scales. Then, each subseries is used to predict the stock price using two types of neural network (NN) models with one and two hidden layers. Results show that having multiple time windows in input datasets together with DWT decrease the RMSE of NN models below 10%.

KW - Apple Stock Prices

KW - Discrete wavelet transform

KW - financial time series data

KW - neural networks

KW - stock price forecasting

UR - http://www.scopus.com/inward/record.url?scp=85060653534&partnerID=8YFLogxK

U2 - 10.1109/UBMK.2018.8566614

DO - 10.1109/UBMK.2018.8566614

M3 - Conference contribution

AN - SCOPUS:85060653534

T3 - UBMK 2018 - 3rd International Conference on Computer Science and Engineering

SP - 518

EP - 521

BT - UBMK 2018 - 3rd International Conference on Computer Science and Engineering

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd International Conference on Computer Science and Engineering, UBMK 2018

Y2 - 20 September 2018 through 23 September 2018

ER -

Kulaglic A, Ustundag BB. Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks. In UBMK 2018 - 3rd International Conference on Computer Science and Engineering. Institute of Electrical and Electronics Engineers Inc. 2018. p. 518-521. 8566614. (UBMK 2018 - 3rd International Conference on Computer Science and Engineering). doi: 10.1109/UBMK.2018.8566614

Stock Price Forecast using Wavelet Transformations in Multiple Time Windows and Neural Networks

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this