As I read about stochastic MPC approaches, there are some approaches claim satisfaction of chance constraints in prediction or in closed-loop sense using different ways of tightening the chance constraints. I am wondering what is the difference between both and how to test them in practice?