Suppose if I have a stochastic process which is a submartingale. What is the "practical" benefit from this property?
I have a roughly idea that this submartingale property suggests a favorable game scenario which is better than the fair game scenario (e.g., if a stochastic process is a martingale, then it can be used to model a fair game). But I'm pretty interested to know if there are any other practical benefits if a stochastic process is a submartingale process?
A submartingale can be decomposed into a martingale and a previsible process