I am trying to implement a strategy that exploits potential misspecifications in density predictions (e.g.: long states with too-low probability; short states with too-high probability).
In particular, I am looking for an option-based strategy that exploits:
The location of the forecast density (i.e.: misspecified mean): Which strategy could be used to benefit from a density prediction that is displaced to the left/right?
The scale of the forecast density (i.e.: missspecified volatility): Which strategy could be used to benefit from a density prediction that exhibit excessive dispersion/concentration?
Asymmetric tail estimates: Which strategy could be used to benefit from densities that assign too-low probability to the left tail compared to the right tail, or vice versa?
Since the strategy concern different payoff regions, I am initially looking at an option-based approach that employs calls and puts with different strikes.