This question relates to credit portfolio analysis. I was asked by my teacher the following question :
Why would a bank use MC simulation in the implementation of the covariance model (a bottom-up model), rather than use the analytical approach? State 2 reasons.
So far I have come up with that the co-variance model requires an estimation of loss correlation and Monte Carlo simulation can be used to generate an infinite number of scenarios of Unexpected losses from historical observations.
I don't know what the second reason for the implementation.