In this talk we first provide a whistle-stop tour of BP Supply & Trading’s commercial activity in the global energy and commodities markets, before giving an overview of the four analytics disciplines
that reside within the trading organization, and quantitative analytics in particular. We then proceed to dive into a single valuation example, that of natural gas storage assets, where we consider how best to determine the value of such a facility from a
financial derivatives perspective, taking into account a stochastic term structure model of the gas price curve, and the engineering constraints of the asset. In doing so we focus our attention on solutions drawn from the field of Stochastic Control and seek to determine admissible exercise policies that maximize the expected financial value. We consider lattice based, Monte Carlo and Quantization methods for this purpose. Throughout our talk there is a strong emphasis on what is required for these models to be practicable in a fast-paced commercial environment, where robust valuations, stable option Greeks and speed of computation are critical. Also
discussed is the challenge presented by sparse data sets for model calibration, and some approaches we employ to assess the impact of transaction costs and hedging frequency on the financial value that might be captured by the commercial teams managing the risk exposure.