<t>Hey GuysI have a question that I hope you can help me with. I have made a model for stock data. More specificaly, I have made a cointegration model, which I have specified pretty well, looking at the relationship between sectoral stock indices from 1926 to 2007. I would then like to test it, and ...

GARCH should be used more often - vol traders in top banks don't use it - instead they take a simple average of the last 6 periods! Oh my G...

<t>I agree - but for practical purposes, BS maybe has a value to add? If the trader is aware of the pitfalls, don't you think he could achieve some info from the BS price? I think other models would be more difficult to implement in practice, although I think it is about time that it starts happenin...

<t>Proper econometrics means applying a solution that makes sense. Fx - don't use OLS on time series if non-stationary (which is what they did), use cointegration instead!Proper econometrics doesn't imply using the toughest solution of all, it just implies not using a false solution! That's what I m...

The first part is because of the statistical properties (nonergodicity), the moving averages is because it has been statistically "proven" that they can't predict anything, and wavelets, honestly, i don't know very well (will get back on that one)

I agree, but what you refer to is "simple stuff" - regressing returns against factors is really not what I would call a very good approach! Neither in the CAPM example. And moving averages are, IMHO, bulls...! Just like any other charting means, like waves, fibonacci stuff etc!

<t>Hi GuysAs the header sounds - who does that? I mean, in banks, quants do programming and modelling, structurers do a little econometrics but not much, and traders certainly don't. Who, except researchers, use that? And who do economic analysis?I know it might sound stupid, but I have never met an...

<t>Hi - I have a small brain teaser for all of youLong in X different UK stocks, all on the FTSE 100 but you don't know which they are, and you don't know the weightings except that they are all together worth 100 mm. Short 100 mm FTSE100 futures (perfect ftse's).The tracking error of the PF is 1%If...

Okay, thanks for the reply - amazing that such an old book can be so expensive!

<t>I'd say (i don't know how many you need)Forecasting, analysis of time, database, corporate, bayesian, computer, linear, dynamic - in that order. I am on trading floor in GS, and I'd say those are the most important, but honestly don't think it matters really much so go with what you like the most...

Bump...

<t>Hi allI'm doing a master in finance next year, and need to do a probability course. I can choose between two, one harder than the other. I would like your help to determine the level of the courses. I have listed the books used, so if anyone has experience with some of them, please let me know th...