New regression-based algorithms for optimal stopping problems reduce computational cost.
problem Optimal stopping problems in dynamic programming.
method Pseudo-regression approach using Monte Carlo approximation of inner products.
result The method asymptotically leads to lower computational cost and complexity.