Skip to content
Projects / Phase 4.3 · JD to Projects

Natural-Gas Storage Valuation, Hedging and LNG Cargo Optionality

What is a storage contract worth, how to run and hedge it, and should the desk buy it?

Open the projectSee all projects

Headline result

Valuation, hedging and a bid-or-pass rule for a 1 Bcf salt-cavern lease, scored on 57 monthly offers from 2019 to 2023. The average lease was valued at $358k and realised $330k; one-day execution lag and unhedged days gave back $877k of $1.03M re-optimisation gains. The bid rule made $10.7M against −$0.6M for always bidding.

Figures

  • Intrinsic and LSMC offer-date values against realised daily-hedged value for every monthly offer, 2010 to 2023.
    Model values form a narrow band while realised outcomes scatter widely, from a $2.1M loss in September 2021 to over $2M.
  • P&L explain waterfall for the mean evaluation lease: intrinsic, re-optimisation, unhedged days and lag, costs, realised.
    Re-optimisation adds $1.03M on paper, but execution lag and unhedged days give back $877k, leaving $330k realised.
  • Realised hedged values per evaluation offer against the model’s 5 to 95% interval, 2019 to 2023.
    Only 74% of outcomes fell inside the model’s 90% interval, against 85% required; the largest misses are late-2023 offers above $3.5M.
  • Realised extrinsic value per offer with stress terms marked, and cumulative profit of three bidding rules.
    Offers struck before Winter Storm Uri (2021) and in 2022 carried 65% of realised extrinsic value; the bid-or-pass rule made $10.7M against −$0.6M for always bidding.

Data used

  • NYMEX natural-gas futures settlements (contracts 1–4) as republished by EIA, 1993 to Apr 2024
  • Henry Hub daily spot from EIA, 1997 to Oct 2026
  • EIA weekly working-gas storage reports, 2010 to Oct 2026

Built from Gas Valuation and Structuring Quant job descriptions

The role asks for

  • Storage and structured deal valuation
  • Monte Carlo and optimisation methods
  • Commodity derivatives pricing and hedging
  • Risk measurement and P&L explain
  • Python
  • Working closely with traders

This project uses

  • Two-factor forward-curve model
  • Least-squares Monte Carlo storage valuation
  • Rolling intrinsic with futures hedging
  • Dual upper bounds
  • P&L explain and VaR backtesting
  • Bid-or-pass deal rule

Members open the full write-up, notebook, tested code and data in the project reader.