James A. Tilley,Valuing American Options in a Path Simulation Model, here. He writes well and the multi factor prose is interesting. This argument is for Trading book valuation use of multi factor models at some level. Probably that is why those guys ran a 12 factor HJM. He ran his simulator in APL in the 90s which then was optimized to some C++ object oriented monstrosity with templates , STL, and God knows what else, which just goes to show bad things can happen to good people?
The goal of this paper is to dispel the prevailing belief that American-style options cannot be valued efficiently in a simulation model, and thus remove what has been considered a major impediment to the use of simulation models for valuing financial instruments. We present a general algorithm for estimating the value of American options on an underlying instrument or index for which the arbitrage-free probability distribution of paths through time can be simulated. The general algo- rithm is tested by an example for which the exact option premium can be determined.
Jarrow and van Deventer, The Arbitrage-free valuation and hedging of demand deposits and credit card loans, 1997, here. Again it is for valuation and risk management rather than ALM simulation for expected balances. The Kamakura stuff on the website is pretty formidable, but it is really for risk management.
Using a market segmentation argument, this paper uses the interest rate derivative’s arbitrage-free methodology to value both demand deposit liabilities and credit card loan balances in markets where deposits/loan rates may be determined under imperfect com- petition. In this context, these ®nancial instruments are shown to be equivalent to a par- ticular interest rate swap, where the principal depends on the past history of market rates. Solutions are obtained which are independent of any particular model for the evo- lution of the term structure of interest rates.