I was working on the pricing of complex american options when I noticed that the exercise is often (very) subobptimal. (If you are not familiar with american options: these are options that can be used on multiple dates, so you have to choose optimally without knowing the future.)
It seems that the clients are more sensitive to past growth or drop in rates than to their value at the moment, wich is a non optimal behaviour. I am looking for a way to modelise the suboptimal behaviour and I tought about Machine learning. But I can't find any reference on suboptimal options exercice. Do you have any broader exemple of Machine learning applied to the replication of human non optimal behaviour ? Do you have any tought on that ?
[link][comment]