The function performs similar tasks as the rational expectations version of Monte Carlo filtering but also uses the steady-state solution under adaptive learning. This means that besides checking the RE solution properties for the prior draws, the function also checks if the steady-state AL solution is convergent when the RE version is unique and convergent.
Page url: http://www.texlips.net/yada/help/index.html?monte_carlo_filtering2.htm