A Haskell implementation of a basic, n-step recombining binomial tree model for pricing American and European options.
- Implement backward dynamic programming algorithm using actual DP (cache results of computation)
- Allow the user to specify call vs. put options
- Allow user to specify European vs. American options
- Maybe generalize to arbitrary payoff structures (at maturity), as specified by some function of type Input -> Double?
- Add support for non-recombining models
- can't do DP, but can solve for price using closed-form equation
- also can solve systems for valuation/replicating strategy at arbitrary time
- Let user specify whether tree is recombining or not, or detect automatically