Serving the Quantitative Finance Community
Least Squares Monte Carlo Degrees of Freedom
I am using the Least Squares Monte Carlo method by Longstaff and Schwartz (2001) to price American Options. When estimating the contnuation value by regression at a specific time-step, what happens if the degrees of freedom (number of coefficients need to be estimated) is larger than the observable (ITM paths as they use)? How does one perform the optimal exercise strategy at this time step?
