Paper: arXiv 2312.12788
Authors: Radhika Prosad Datta
Abstract
This paper explores the application of Sample Entropy (SampEn) as a sophisticated tool for quantifying and predicting volatility in international oil price returns. SampEn, known for its ability to capture underlying patterns and predict periods of heightened volatility, is compared with traditional measures like standard deviation. The study utilizes a comprehensive dataset spanning 27 years (1986-2023) and employs both time series regression and machine learning methods. Results indicate SampEn’s efficacy in predicting traditional volatility measures, with machine learning algorithms outperforming standard regression techniques during financial crises. The findings underscore SampEn’s potential as a valuable tool for risk assessment and decision-making in the realm of oil price investments.
Complexity vs Empirical Score
- Math Complexity: 4.5/10
- Empirical Rigor: 7.0/10
- Quadrant: Street Traders — empirical and applied, lower math complexity
Why this score: The paper applies a well-defined metric (Sample Entropy) with established formulas rather than developing new complex mathematics, but implements it across a comprehensive 27-year dataset using multiple ML techniques (SVM, KNN) and time series regression with specific ARIMA(4,1,3) parameters, demonstrating practical implementation and performance evaluation.
Research Flowchart
flowchart TD
A["Research Goal: Apply Sample Entropy<br>to predict oil price volatility"] --> B{"Data Preparation"}
B --> C["Dataset: 27 Years<br>International Oil Prices 1986-2023"]
C --> D["Feature Engineering"]
D --> E["Compute Sample Entropy<br>Standard Deviation<br>Lagged Returns"]
E --> F{"Computational Models"}
F --> G["Time Series Regression"]
F --> H["Machine Learning Models"]
G --> I["Findings:<br>SampEn effectively predicts<br>traditional volatility"]
H --> J["Findings:<br>ML outperforms regression<br>during financial crises"]
I --> K["Final Outcome:<br>SampEn is a valuable tool<br>for risk assessment &<br>oil price investment decisions"]
J --> K
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes using Sample Entropy as a tool to quantify and predict volatility in international oil price returns.
Source sentence
This paper explores the application of Sample Entropy (SampEn) as a sophisticated tool for quantifying and predicting volatility in international oil price returns.
Assumes 2
- The time series of rolling‑window standard deviation and Sample Entropy are stationary. statistical
Source sentence
ADF Statistic: -4.503642 , p value 0.01 which rejects the null hypothesis i.e. we conclude that the time series ts_std is stationary.
- The two stationary series are autocorrelated, preventing simple linear regression. statistical
Source sentence
The results of the Ljung Box test for ts_std gives a p-value < 2.2e-16 which implies that the null hypothesis can’t be rejected i.e. the series is correlated.
Data
- Crude oil price series (DCOILWTICO) market-prices public
Source sentence
The data consists of the price of crude oil as reported by the Federal Reserve Economic Data (FRED) and is available at https://fred.stlouisfed.org/series/DCOILWTICO. We consider the starting point of our data from 2nd January 1986 to 10th April 2023 a total of 9389 points.
Costs
Not applicable
Finds 2
- Sample Entropy effectively predicts traditional volatility measures, and machine learning models outperform standard regression during financial crises. predictive-accuracy
Source sentence
Results indicate SampEn's efficacy in predicting traditional volatility measures, with machine learning algorithms outperforming standard regression techniques during financial crises.
- The Pearson correlation between rolling‑window standard deviation and Sample Entropy is -0.704, indicating a significant negative relationship. statistical
Source sentence
... is calculated to be -0.704 which shows that there is a significant negative correlation between the 2 time series.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 18 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.