ARIMA for short-term and LSTM for long-term in daily Bitcoin price prediction

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F70883521%3A28140%2F23%3A63570219" target="_blank" >RIV/70883521:28140/23:63570219 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-031-23492-7_12" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-23492-7_12</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-23492-7_12" target="_blank" >10.1007/978-3-031-23492-7_12</a>

Jazyk výsledku

angličtina

Název v původním jazyce

ARIMA for short-term and LSTM for long-term in daily Bitcoin price prediction

Popis výsledku v původním jazyce

The goal of this paper is the insight into the forecasting of Bitcoin price using machine learning models like AutoRegressive Integrated Moving Average (ARIMA), Support vector machines (SVM), hybrid ARIMA-SVM, and Long short-term memory (LSTM). Depending on the different types of data and the period, various models are used for prediction. A single model may be the best fit in the short term but may not be the best in long-term series data. Thus, using only a single model may not be suitable for forecasting time series data that depends on data sampling length and prediction time, and the type of specific applications. As a result, the ARIMA model produces better error results with a short prediction period or a small data set. In contrast, the Hybrid ARIMA-SVM model will help improve the performance of the ARIMA model when predicting over a long period, specifically 7 and 30 days for Bitcoin price prediction used in this research paper. The paper aims to compare traditional models such as the ARIMA, the Hybrid ARIMA-SVM, and deep learning models such as LSTM on a specific cryptocurrency prediction task using different scenarios. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Název v anglickém jazyce

ARIMA for short-term and LSTM for long-term in daily Bitcoin price prediction

Popis výsledku anglicky

The goal of this paper is the insight into the forecasting of Bitcoin price using machine learning models like AutoRegressive Integrated Moving Average (ARIMA), Support vector machines (SVM), hybrid ARIMA-SVM, and Long short-term memory (LSTM). Depending on the different types of data and the period, various models are used for prediction. A single model may be the best fit in the short term but may not be the best in long-term series data. Thus, using only a single model may not be suitable for forecasting time series data that depends on data sampling length and prediction time, and the type of specific applications. As a result, the ARIMA model produces better error results with a short prediction period or a small data set. In contrast, the Hybrid ARIMA-SVM model will help improve the performance of the ARIMA model when predicting over a long period, specifically 7 and 30 days for Bitcoin price prediction used in this research paper. The paper aims to compare traditional models such as the ARIMA, the Hybrid ARIMA-SVM, and deep learning models such as LSTM on a specific cryptocurrency prediction task using different scenarios. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, ICAISC 2022, PT I

ISBN

—

ISSN

2945-9133

e-ISSN

1611-3349

Počet stran výsledku

13

Strana od-do

131-143

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Zakopane

Datum konání akce

19. 6. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000972696000012