Serving the Quantitative Finance Community

 
User avatar
rafaelvivo
Topic Author
Posts: 0
Joined: September 21st, 2007, 11:13 pm

Gas Storage Valuation

March 12th, 2008, 9:28 am

Hi! Im doing a dissertation about gas storage valuation. I have some problems, if anybody believes that can help me and discuss about it, please answer
 
User avatar
PaperCut
Posts: 0
Joined: May 14th, 2004, 6:45 pm

Gas Storage Valuation

March 12th, 2008, 2:12 pm

QuoteOriginally posted by: rafaelvivoHi! Im doing a dissertation about gas storage valuation. I have some problems, if anybody believes that can help me and discuss about it, please answerStoring Arb.
 
User avatar
rafaelvivo
Topic Author
Posts: 0
Joined: September 21st, 2007, 11:13 pm

Gas Storage Valuation

March 13th, 2008, 10:09 am

Hi outrun,I´ve done the valuation under multifactors spot models. But now, I want to valuate the storage contract under the "rolling intrinsic" strategy, and I dont know how to do it. If you know something about it, tell me please. I have problems in both simulate multifactor forward curve price and to apply the optimization
 
User avatar
rafaelvivo
Topic Author
Posts: 0
Joined: September 21st, 2007, 11:13 pm

Gas Storage Valuation

March 13th, 2008, 10:11 am

Thank you papercut, but I´ve readed this article. I have problems in putting into practice the simulation and the optimization. Are there Matlab code for simulate multifactor forward curve?
 
User avatar
rafaelvivo
Topic Author
Posts: 0
Joined: September 21st, 2007, 11:13 pm

Gas Storage Valuation

March 13th, 2008, 11:34 am

Thank you so much for you help, outrun (by the way, nice play the outrun. Yesterday i played in a Mame emulator)I guess you are talking about the intrinsic value of a contract. That is, the value of storage with the current forward curve (fully-hedged strategy). I have programmed this issue (with backward induction), but I think I will check the value under your linnear programming explanation.But, indeed I want to do the following:Rolling intrinsic is a strategy that can create extrinsic value (join with spot based models). First day of the storage contract, storage operator value the storage under the current forward curve (like you explained). But nex day of the contract, the operator check the new market forward curve. If with the new market forward curve you make a profit, then the operator unwind yesterday positions and take the new positions (offsetting transaction costs), and finally update the storage value of yesterday with this new profit. So, during the life of the contract, because of the evolution of the forward curve movements, you can make new profits unwinding and taking new positions. The total value of the storage is the sum of the initial value (intrinsic value) and this profits, called extrinsic value.So, to calculate this profits, you must do two things (and I dont know how to do this two things):1.- Simulate the evolutions of the forward curve during the year of the storage contract.2.- Take the optimal position under the new forward curve and calculate the net profit, during the year, involving dynamic programming.
 
User avatar
tristram
Posts: 0
Joined: March 13th, 2008, 3:33 pm

Gas Storage Valuation

March 17th, 2008, 9:43 am

QuoteOriginally posted by: outrunQuote2.- Take the optimal position under the new forward curve and calculate the net profit, during the year, involving dynamic programming.Its complicated. Suppose you have a flat forward curve with the price of the last month a tiny bit higher than all other months. With dynamic programming & estimating the 'static intrinsic value', the optimal solution would be to fill it up in the initial months & sell the whole volume in the last most. Suppose that with that schedule a trader trades the forward markets & locks the intrinsic value.Now suppose that the different between the last months and all other increases..The trader won't be able to do a thing, because it has already hedged the intrinsic optimal schedule, making a tiny profit. The optimal intrinsic schedule doesn't change if the price difference increases, so the trader can't do a trade. He should have done stochastic optimization, looking at all possible future forwardcurves, and make a decision based on uncertain future curves instead of the intrinsic schedule.That is a good example. There are several approaches adopted for solving this type of problem, but my preferred method is stochastic DP. Whatever approach you choose, if it doesn't consider the stochastic nature at every decision point, then it will be missing something.What you gain with a rolling intrinsic valuation is certainty of profit at each point in time. What you lose, though, is possibility of greater profits, as demonstrated in the example provided by outrun. Remember that on the day, gas will be traded at spot, regardless of what forward deals you have in place. On the day, purchase from spot to meet your forward obligations. Also on the day, if the spot price is higher than your expected marginal value of gas in storage, sell some gas from storage. Transaction costs aside, these are two separate activities, which just happen to be in the same market.Put slightly differently, you can undertake a rolling intrinsic type of strategy without ever having any gas storage available to you. If you choose to use physical (or financial) storage to provide certainty in this strategy, that is fine, but it is not required, and you should realise that there is an opportunity cost for doing so. If you didn't use your storage space for hedging the rolling intrinsic deals, you would instead be able to use it for (hopefully) profitable trading against spot, hence the opportunity cost.