Attended this research workshop yesterday by a fellow PhD student, Kapala Srinivas Rao. The abstract of the topis is here:

The Natural gas industry in US was regulated by regulatory agencies since 1930s.This regulation created distortion in natural markets and gas shortages were resulted in 1970s particularly in interstate markets of the natural gas industry. The gas shortages prompted the deregulation of the industry in order to promote efficiency in the industry. The process was launched by adoption of Natural Gas Act 1978 which partially liberalized the interstate markets. The complete deregulation of markets followed when FERC adopted order No.436 of 1985 and order No.636 of 1992 which introduced open access to interstate pipeline transportation, allowed gas marketing companies and unbundled natural gas sales from pipe line transportation. The deregulation has changed the structure of industry and has resulted in spot market at the wholesale level (Whole gas market). As spot gas prices become highly volatile in the open markets, participants in the spot markets face substantial price risks. So, the financial gas market had its beginning in 1980s and developed by 1990s to hedge this price risk.

Under deregulated environment, the pricing of the natural gas contracts has become an important issue since they have to be priced according to the financial risk. Contracts in natural gas markets usually are designed to provide flexibility of delivery with respect to both the timing and the amount of natural gas due to complex patterns of consumption of gas. These contracts are called as Swing contracts. Swings permit the option holder to repeatedly exercise the right to receive greater or smaller amounts of energy subject to daily as well as periodic constraints.

Researchers have suggested Binomial/Trinomial tree approach and Binomial/Trinomial forest approach (a multiple layer tree extension of the binomial/trinomial tree approach) to price swing contracts but the contracts are assumed to be simple and single period type. However, swing contracts are usually of multi-period type with additional provisions. So I will extend the Binomial/trinomial forest model suggested in the literature to price multi-period contracts with an additional provision called as make-up provision. There are three steps in pricing these contracts. The first step consists of empirical estimation of parameters of stochastic process that is assumed to be followed by underlying commodity price i.e. spot price of the Natural gas from the market data (NYMEX Futures data). The second step consists of building the trinomial trees to approximate the stochastic process and the last step consists of writing an algorithm for pricing a swing option. In the presentation, I will be emphasizing on the last step i.e. how the algorithm works.