Application of Markov Model in Crude Oil Price Forecasting

Isah, Nuhu; Bon, Abdul Talib

doi:10.22178/pos.25-3

Cited by 5 publications

(4 citation statements)

References 11 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Statistical models, which are also known as random time series models, include exponential smoothing (ES) (see, e.g., Kourentzes et al [13]), auto-regressive integrated moving average (ARIMA) model (see, e.g., Guo [14]), generalized auto-regressive conditional heteroskedasticity (GARCH) model (see, e.g., Zhang et al [15]), hidden Markov model (HMM) (see, e.g., Isah & Bon [16]), and vectorial auto-regression (VAR) (see, e.g., Mirmirani & Li [17]). For example, Zolfaghari & Gholami [18] showed that ARIMA models had a good forecasting impact on international crude oil prices.…”

Section: Forecasting By Statistical Modelsmentioning

confidence: 99%

A New Forecasting Approach for Oil Price Using the Recursive Decomposition–Reconstruction–Ensemble Method with Complexity Traits

Wang

2023

Entropy

View full text Add to dashboard Cite

The subject of oil price forecasting has obtained an incredible amount of interest from academics and policymakers in recent years due to the widespread impact that it has on various economic fields and markets. Thus, a novel method based on decomposition–reconstruction–ensemble for crude oil price forecasting is proposed. Based on the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) technique, in this paper we construct a recursive CEEMDAN decomposition–reconstruction–ensemble model considering the complexity traits of crude oil data. In this model, the steps of mode reconstruction, component prediction, and ensemble prediction are driven by complexity traits. For illustration and verification purposes, the West Texas Intermediate (WTI) and Brent crude oil spot prices are used as the sample data. The empirical result demonstrates that the proposed model has better prediction performance than the benchmark models. Thus, the proposed recursive CEEMDAN decomposition–reconstruction–ensemble model can be an effective tool to forecast oil price in the future.

show abstract

Section: Forecasting By Statistical Modelsmentioning

confidence: 99%

A New Forecasting Approach for Oil Price Using the Recursive Decomposition–Reconstruction–Ensemble Method with Complexity Traits

Wang

2023

Entropy

View full text Add to dashboard Cite

show abstract

“…In addition, these models do not address the problem of dependency or correlation in data. The use of the hidden Markov models (HMMs) in forecasting the crude oil prices was adopted by several authors, for example, Bon and Isah (2016) and Isah and Bon (2017), who have both used, and also other researchers interested in using the HMMs in analyzing the oil prices, the frequentist principle in estimating and forecasting the model. Gong et al (2020) used a five-variable Markov switching vector autoregression (Markov switching VAR) taking into account the following variables: oil prices, oil aggregate demand, oil aggregate supply, global oil inventory and oil speculative demand.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Modelling of crude oil price data using hidden Markov model

Kadhem

Thajel

2023

JRF

View full text Add to dashboard Cite

PurposeOne of the most important sources of energy in the world, due to its great impact on the global economy, is the crude oil. Due to the instability of oil prices which exhibit extreme fluctuations during periods of different times of market uncertainty, it became hard to the governments to predict accurately the prices of crude oil in order to build their financial budgets. Therefore, this study aims to analyse and model crude oil price using the hidden Markov process (HMM).Design/methodology/approachTraditional mathematical approaches of time series may be not give accurate results to measure and analyse the crude oil price, since the latter has an unstable and fluctuating nature, hence, its prediction forms a challenge task. A novel methodology that is so-called the HMM is proposed that takes into account the heterogeneity in prices as well as their hidden state-based behaviour.FindingsUsing the Bayesian approach, several estimated models with different ranks are fitted to a non-homogeneous data of Iraqi crude oil prices from January 2010 into December 2021. The model selection criteria and measures of the prediction performance of each model are applied to choose the best model. Movements of crude oil prices exhibit extreme fluctuations during periods of different times of market uncertainty. The processes of model estimation and the model selection were conducted in Python V.3.10, and it is available from the first author on request.Originality/valueUsing the Bayesian approach, several estimated models with different ranks are fitted to a non-homogeneous data of Iraqi crude oil prices from January 2010 to December 2021.

show abstract

“…Oil and gas play a momentous role in Malaysia, PETRONAS is the country's owned oil and gas company that contributed immensely to the economic development of Malaysia. PETRONAS contributed in the education sector of Malaysia; it also helped to the annual budget of Malaysia almost every year [4].…”

Section: Introductionmentioning

confidence: 99%

A Research on Markov Model and Hidden Markov Model in Crude Oil Price Forecasting for PETRONAS Malaysia

2019

IJITEE

View full text Add to dashboard Cite

Crude oil price forecasting is an essential component of sustainable development of many countries as crude oil is an unavoidable product that exists on earth. In this paper, a model based on a hidden Markov model and Markov model for crude oil price forecasting was developed, and their relative performance was compared. Path analysis of Structural Equation Modelling was employed to model the effects of forecasted prices and the actual crude oil price to get the most accurate forecast. The key variables used to develop the models were monthly crude oil prices s from PETRONAS Malaysia. It was found that the hidden Markov model was more accurate than the Markov model in forecasting the crude oil price. The findings of this study show that the hidden Markov model is a potentially promising method of crude oil price forecasting that merit further study.

show abstract

Application of Markov Model in Crude Oil Price Forecasting

Cited by 5 publications

References 11 publications

A New Forecasting Approach for Oil Price Using the Recursive Decomposition–Reconstruction–Ensemble Method with Complexity Traits

A New Forecasting Approach for Oil Price Using the Recursive Decomposition–Reconstruction–Ensemble Method with Complexity Traits

Modelling of crude oil price data using hidden Markov model

A Research on Markov Model and Hidden Markov Model in Crude Oil Price Forecasting for PETRONAS Malaysia

Contact Info

Product

Resources

About