ECTAP
 
HomeDespre ECTAEventsPolitica editorialaTrimite un articolParteneri / link-uri utileArchiveAbonamentContact
 

ISSN 1841-8678   (print)
ISSN 1844-0029   (online)

News

Archive ECTAP

Note: for the period 1994-2003 the archive of the magazine will not be available online

Supplements ECTAP

If you cannot open the pdf file you need Adobe Reader.
download Adobe Reader

Creative Commons License

Theoretical and Applied Economics
No. 2 / 2020 (623), Summer

Analyzing the robustness of ARIMA and neural networks as a predictive model of crude oil prices

Sudhi SHARMA
Apeejay School of Management, Dwarka, New Delhi, India
Miklesh YADAV
FIIB, Business School, New Delhi, India

Abstract. The paper is focusing in analyzing the robustness of the Auto Regressive Integrated Moving Average (ARIMA) and Artificial Neural Networks (ANNs) as a predictive model in forecasting the crude oil price. The paper has identified stochastic trend in the daily time series data starting from (03.01.2011 to 11.10.2019). The time considered in the study is subject to high volatility, which makes this paper unique from the current stock of knowledge. During this time frame it has been identified that there is no structural break. The empirical analysis furnishes that the ARIMA is the best suited model. The decision criterion for the selection of the best suited model depends on ME, RMSE, MAE and MASE. From the results of the criterion it has found that both the models are providing almost closed results but again ARIMA is the best suited model for the current data set.

Keywords: ARIMA, ANNs, Crude-Oil.

Download the full article:  

Contents

Open acces

ECTAP

Search

BOOKS

The Economicity. The Epistemic Landscape, Marin Dinu, 2016

Partners


ISSN 1841-8678 (ediția print) / ISSN 1844-0029 (ediția online)
© Copyright Asociația Generală a Economiștilor din România (AGER) / General Association of Economists From Romania  (GAER)
Redacția: 010702, București, Calea Griviței nr. 21, sector 1, E-mail: economia.ta@edeconomica.com

© 2006-2020 AGER