Your question is Validate a Loss Model Without Data. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
If you have a loss model with no historical data to back it up and no benchmark data, how do you validate that model?