<t>Obviously, Alan, in the short run I completely agree with you - as I implied. But I still think it's dangerous to bake price level relationships into dynamic models, at least unless you are being much more precise and specific about the sources of the dependence than a crude "how log-normal is it...

<t>You want to be a little careful about how you handle non log-normality for price processes. The reason why much of the literature about backbones etc is for interest-rates is that they are not prices. In the long-run, you should not expect, e.g., a gas price to look very different proportionally ...

<t>I strongly agree with outrun, by the way: in my experience, even if it's constant, the model is correctly specified, and you have a decent amount of data, you often don't get very precise estimates of quantities like mean-reversion rates. Even if it's a deterministic function of observables (like...

If you've specified the process by which the mean-reversion rate evolves you can try to estimate the whole thing on the full sample by max likelihood or MCMC.

<t>Plus, another issue that the monetarists found in the eighties: the relationship between money supply and inflation is very unstable. It would be rather hard to calibrate your algorithm's response to inflation signals.And this is even before you consider that you might want to include other varia...

<t>Ha! Well, since it would be Hayek money, "namely the best money ever devised", what could possibly go wrong? But this still doesn't really solve the problem: which commodities go in your basket? And how does the automated system know when to accommodate price shocks rather than to resist them? e....

<t>There is not currently a well defined algorithm for currency stability. People have made claims about, e.g., the Taylor rule, but monetary policy is still complex and discretionary enough that I've never seen any serious candidates for fully automatic policy.Incidentally, all this recent chatter ...

<t>Thank you, Engy, I did read the paper.Most of us don't have the luxury of only being judged on terminal PL; all too often, vega effects are terminal in quite another sense. By all means, do have this conversation with your boss and you risk-manager next time you have a large drawdown :)Yes, there...

And, also, these are American options: put/call parity doesn't hold...

<t>This is a fairly useful exercise, but is still simulated. Worth also looking at actual backtests - particularly because it looks like this paper has fixed vols and no vega effects, which are rather important.You'll notice that, if you cast a quick eye over the time-based hedging tables at the end...

<t>It is definitely worth trying rules-based hedges in backtests, though you will need tick-data if you're going to look at intraday problems, which complicates things a bit, particularly for swing and storage instruments. For exotics, you have to model the instrument price-change, and that means yo...

<r>Oh, Lord, no. I'm certainly not saying the model doesn't matter. The model matters a lot. All I meant is that your other modeling choices are going to dominate choosing whether to use "r - q" or "r + lambda - q" in your drift specification, for example. You will need the "lambda" if you are estim...

<r>I'm not saying at all that you should assume the parameter is zero. I'm trying to explain why worrying overly at this stage about which measure you will use is a little beside the point, which was your original question. As I posted earlier, I agree with bearish's point about simultaneously estim...

QuoteOriginally posted by: yyangThere was a bug in my timestamp parser. Please bury me alive.Happens to us all from time-to-time; I just hope you didn't put money on it

<t>Yes - extrinsic value is another way to describe it.In another few hints as to why which measure you use is not your biggest problem (and also, indirectly, why the answer is hard to answer even in principle), think about this:For Natural Gas (and I'm only going to talk about US-traded instruments...