January 27th, 2014, 5:14 pm
I don't know those markets, so can only make general comments. It might make sense to suppose you have a non-tradable (or at least non-storable) spot, underlying some futures that do trade.Let's call the spot process [$]X_t[$]; since it cannot be practically held I am guessing this allows [$]X_t[$] to move in a mean-reverting way between two barriers (f,c) for floor and cap without an arbitrage opp. (Is that correct -- it's the inability to store that prevents an arbitrage on a barrier hit?)Forward prices are given, under some market pricing process by [$]F_{t,T} = E_t[X_T][$]. Since the support of X is in (f,c), you will always find the forwards in the same interval.Now a good model will identify some good state variables.Putting aside the issue of good state variables, the nature of the barriers matters for your question. Does the spot sometimes reach one of the barriers? If so, what happens next?
Last edited by
Alan on January 26th, 2014, 11:00 pm, edited 1 time in total.