<t>I was pleasantly surprised to read some work on the philosophy of economics last month, which shows the progress being made to put economics on a more sound philosophical basis - something that hasn't been properly done before and the lack of which is a source of many of the problems that cause P...

<t>Let's get the problem properly expressed. You don't want to fit the data, you want to get a good forward looking estimate of volatility for some task. Which is the best model for estimating volatility will depend on the task you want to do. Change the task and you may have to change your forecast...

<t>What a bloody waste of time it was following this thread. I thought we might get somewhere, but N just refused to say what he was talking about in a manner that others could understand. And it's not because I don't get the maths, physics and all the rest that N was spouting. I do understand, but ...

Pertinent to this forum is does it have any relevance for finance. I think not. Wake me when it does.

KackToodles,You misunderstood my point. I very carefully said there is no "market trading strategy" that can be exploited without limit. I agree with your points that markets themsleves add value in many instances.There's a big difference.

<t>At least with the physical world we know (or strongly suspect) that there is something real out there independent of us. Markets are competely different - they're social constructs and self-referential in a way that doesn't fit within any physical theory. As an example, if somebody comes up with ...

Yeah, it's easy. The distribution is everywhere except my farm!

Try it out.

Suck it and see! Perhaps the covariance between the fund and heding instruments isn't too bad, in which case the actual choice of hedge won't really matter. Historical simulation augmented by some fat tails scenarios might show you can get away with a simple robust regression.

Not enough adjectives. There's only so much you can do with "++" and "--".

Why not? Us quants are one big happy anglophone community, only divided by our common language.

You're going to hedge on the results of a model where the parameter uncertainties are huge? Good luck!

What are you going to do with the model when you have it? That determines the method used. But I suspect that if the OLS results are meaningless then so will be any interpretation of the results from any analysis. This is an overhyped area.

<t>It sounds like you want to predict here. In which case you should be testing how well the prediction works, say based on AR(1) or an AR(3) series. Look at how close you came to prediciting the actual number divided by the predicted uncertainty of your estimate. For smoothness/robustness you can u...

"serie" is not a word. "series" is not always a plural, despite the "s" at the end.