Paper: arXiv 2309.13096
Authors: Sarit Maitra, Vivek Mishra, Sukanya Kundu, Manav Chopra
Abstract
This work adopts a novel approach to determine the risk and return of crude oil stocks by employing Arbitrage Pricing Theory (APT) and Quantile Regression (QR).The APT identifies the underlying risk factors likely to impact crude oil returns.Subsequently, QR estimates the relationship between the factors and the returns across different quantiles of the distribution. The West Texas Intermediate (WTI) crude oil price is used in this study as a benchmark for crude oil prices. WTI price fluctuations can have a significant impact on the performance of crude oil stocks and, subsequently, the global economy.To determine the proposed models stability, various statistical measures are used in this study.The results show that changes in WTI returns can have varying effects depending on market conditions and levels of volatility. The study highlights the impact of structural discontinuities on returns, which can be caused by changes in the global economy and the demand for crude oil.The inclusion of pandemic, geopolitical, and inflation-related explanatory variables add uniqueness to this study as it considers current global events that can affect crude oil returns.Findings show that the key factors that pose major risks to returns are industrial production, inflation, the global price of energy, the shape of the yield curve, and global economic policy uncertainty.This implies that while making investing decisions in WTI futures, investors should pay particular attention to these elements
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 6.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced econometric methods like Quantile Regression and Arbitrage Pricing Theory, involving sophisticated statistical modeling and distributional analysis, which raises math complexity. Empirical rigor is moderate with robust statistical validation on real-world WTI data and discussion of market variables, though it lacks explicit backtesting or code.
Research Flowchart
flowchart TD
A["Research Goal: Identify Risk Factors Driving Crude Oil<br>WTI Returns using APT & Quantile Regression"] --> B["Data Collection & Inputs"]
B --> C{"Key Methodology Steps"}
C --> D["Arbitrage Pricing Theory APT<br>Identify Underlying Risk Factors"]
C --> E["Quantile Regression QR<br>Estimate Relationships Across Distribution Quantiles"]
D & E --> F["Computational Process<br>Estimation & Model Stability Analysis"]
F --> G["Key Findings & Outcomes<br>Identifies Major Risk Factors: Industrial Production, Inflation,<br>Energy Prices, Yield Curve, Policy Uncertainty<br>Highlights Impact of Structural Discontinuities"]
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 implements Arbitrage Pricing Theory to identify underlying risk factors and applies Quantile Regression to estimate their effects across the distribution of WTI returns.
Source sentence
The two main contributions to this paper are: ● implementation of APT to identify the underlying risk factors driving ROC, ● application of QR to estimate the effect of these risk factors on different segments of the distribution of WTI returns.
Assumes 6
- Risk factor sensitivities (betas) are estimated using linear regression. model
Source sentence
The βs are estimated by using linear regression.
- Series are differenced to achieve stationarity. statistical
Source sentence
‘D’ prefix added to the relevant dataset after differencing to stabilize the series.
- Errors are heteroscedastic, assessed with Breusch‑Pagan test. statistical
Source sentence
Heteroscedasticity was assessed using the Breusch-Pagan test.
- No autocorrelation in residuals, tested with Durbin‑Watson and Breusch‑Godfrey tests. statistical
Source sentence
the Durbin-Watson (DW) test result of 1.98 indicates that there is no autocorrelation.
Show 2 more
- Quantile regression is appropriate because return distributions are leptokurtic. statistical
Source sentence
Descriptive statistics show relevant distributions are leptokurtic, justifying quantile regression to detect herding bias in tails.
- The linear APT model assumes expected return is a linear function of factor exposures plus an idiosyncratic risk premium. model
Source sentence
The fundamental tenet of APT is that an asset's expected return is a linear function of how exposed it is to different risk factors plus a particular risk premium (idiosyncratic risk) that is unique to that asset.
Data
- FRED macroeconomic monthly data (5 years) macro public
Source sentence
The study used five years of monthly data from the FRED Economic Database to estimate betas, using yield data for crude oil, specifically the WTI crude oil spot price...
- WTI crude oil spot price market-prices public
Source sentence
using yield data for crude oil, specifically the WTI crude oil spot price, to provide insights.
- S&P 500 index (SPY) benchmark public
Source sentence
The S&P 500 index, represented by the SPY, was chosen as the benchmark for the US equity market...
- Macro variables (DGS3MO, DGS5, INDPRO, CPIAUCSL, UNRATE, M1SL, CCU, VIX, GPR) macro public
Source sentence
Table 2 presents selected variables, macroeconomic indicators, and other relevant data sources for the research question.
Costs
Not applicable
Finds 5
- Market return (SP 500) has a negative effect on crude oil returns at the median and 90th quantiles, but not at lower or higher quantiles. statistical
Source sentence
the market return (erSP) has a negative effect on crude oil returns at the median and 90th quantiles, but not at the lower or higher quantiles.
- Industrial production, global economic policy uncertainty, and treasury yield curve have positive effects on crude oil returns across all quantiles. statistical
Source sentence
Production (dPROD), global economic policy uncertainty (dGPE), and the treasury yield curve (dSPREAD) all have positive effects on crude oil returns across all quantiles.
- Money supply (dMONEY) has a large negative effect on crude oil returns at the 25th quantile. statistical
Source sentence
the money supply (dMONEY) has a large negative effect on crude oil returns at the 25th quantile.
- Unemployment rate (dUNRATRE) shows a positive but not statistically significant effect on WTI returns at all quantiles. statistical
Source sentence
dUNRATRE has a positive effect on WTI returns, but not significant at any of the quantiles (Qn 0.25, Qn 0.5, Qn 0.75, and Qn 0.9).
- Inflation (dINFLATION) has a considerable positive effect on crude oil returns at the 25th and 50th quantiles, but not at higher quantiles. statistical
Source sentence
dINFLATION has a considerable positive effect on crude oil returns at the 25th and 50th quantiles, but not at higher quantiles.
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.