Figured it'd be some fun light reading. And how would one make a better risk model for all those < 3 Sigma events?
I'll save you a lot of work. If you can't sleep at night, cut to ½ or ¼ your current size. Your gut knows when you're overextended.
Of course, but perhaps group expertise can cull GOOGLE down to some manageable selections. Otherwise, why forum? Knowledge usually helps inform your gut, no? The uneducated might fear something benign or be unaware of danger. I find it hard to believe that you make any decision solely on guts.
In real life, disastrous losses on complicated derivatives portfolios come either from unanticipated liquidity shortfalls or from second- or third- order effects that are ahistoric and don't show up in risk models. Fancy risk models are just eyewash for the boss or client. An experienced trader knows when his portfolio is too big and unwieldy.