Doctoral thesis

Dynamic Models of Oil Forward Curves

The Oil Forward Curve Risk Premium’s Dynamic: A Practitioner’s Approach.

I follow the methodology in Cortazar et al. (2022). I first validate their approach and then ...

Author(s) :

Xavier Fixaris, PhD

Gasoline Trader, Total Trading SA, Switzerland

Abstract :

The Oil Forward Curve Risk Premium’s Dynamic: A Practitioner’s Approach.

I follow the methodology in Cortazar et al. (2022). I first validate their approach and then extend it by considering both Brent and West Texas Intermediate (WTI) futures, and incorporating two independent sources of forecasts (Reuters and Bloomberg). I empirically test explanatory variables for oil’s risk premia, shedding light on its determinants with additional regressors. I confirm that risk premia are stochastic, and more volatile in the short-term. I also confirm the relevance of equities markets, interest rates, dollar index, hedging pressure and inventories to explain oil risk premia. For inventories, I find that their absolute level rather than their variation is relevant. In contrast to Cortazar et al. (2022), I find no link between the default premium and oil risk premia. I add refining margins as a regressor and demonstrate its relevance. I also find that the oil volatility index (OVX) correlates strongly with the oil risk premia, which raises questions about the choice of model used. I find that oil volatility is correlated with the state variables, while the model assumes the variance of the spot price to be independant from them.

Keywords: Oil, Forward curve, Stochastic factor model, Risk premia, Kalman filter


Which Normality for Oil Prices?

Log-normality of prices and the Black model are the workhorses of oil derivatives pricing and risk management. West Texas Intermediate (WTI) prices going negative in April 2020 shook the trading community and called for a revision of those assumptions. Shortly before this happened, Cortazar et al. (2022) modeled WTI forward curve in order to estimate dynamically the risk premium. In this paper I propose a modification of their approach, using a six-latent-variables model describing jointly several crude benchmarks. I leave aside the dynamic of the risk premium to focus on the forward curve. In this paper, negative prices are allowed, and I propose a unified framework to price options, calendar spread options, and WTI-Brent options.This methodology unifies the pricing framework for a large range of oil derivatives, which makes it useful for risk-management purposes.

Keywords: Oil, Forward curve, Stochastic factor model, Kalman filter, Option pricing, Real options, Risk management

Date : 30/03/2022
Thesis Committee :

Supervisor: Raman Uppal, EDHEC Business School 

External reviewer: Bryan R. Routledge, Carnegie Mellon University  

Other committee members: Emmanuel Jurczenko and Enrique Schroth, EDHEC Business School

See Also

- 27-04-2022
The fifth EDHEC PhD IN FINANCE FORUM will be held online on 7 June 2022. Discover the...
- 08-03-2022
Vijay Vaidyanathan, a Californian alumnus of EDHEC Business School’s PhD in Finance,...
Professor Gianfrate, contributing scientist of the IPCC report on climate change
- 02-03-2022
The Intergovernmental Panel on Climate Change (IPCC) released its latest Climate Report...
What were EDHEC-risk’s top 10 most read articles in 2021?
- 23-12-2021
A year in research: focus on retirement investing (goal-based investing,...