On-line Quantitative Sources

[elliot.wave]

Hi there,

To all the quants out there: can anyone recommend a good on-line source for non-linear time series analysis?

Thanks,

EW

 

[MFE]

www.nonlineartimeseriesanalysis.com

 

[JoeyDVivre]

Are you kidding? That must be a joke

 

[JoeyDVivre]

Anyway, I'll bet you're up to no good....

 

[elliot.wave]

Joey's right - the url is not valid

And I assure you that my intentions are noble... seriously

EW

 

[JoeyDVivre]

So what are your intentions?

 

[Kiwi]

Sorry I forgot to water my quant farm. Google search though provides the following article
http://www.dwyerecon.com/pdf/lectnlin.pdf

I will now provide an executive summary of quotations:
This is a preliminary, very brief summary of nonlinear time series useful for
finance. ...In what follows, I probably would run out of symbols if I tried to use unique symbols for every different function. At the least, I would end up using some unusual letters.

Nonlinear time series often is confused with chaos,

 

[elliot.wave]

Thank you, Kiwi - I will definitely look through it...

EW

 

[JoeyDVivre]

This is such a bad idea.....

 

[elliot.wave]

Joey,

My apologies for being a capital market heretic!

 

[Kiwi]

Joey's right, I'm out of my depth in this area. I would skim it at best as it may certainly not be the best into, to its credit it is <30 pages.

Now I have a Q, JDV, when you think Multivariate Stats, does any particular book or author come to mind? I really need to take some classes but nothings on offer untill next year. Copula's seem to be all the rage...

 

[JoeyDVivre]

Copulas are a really good start when we think that things aren't multivariate normal.

My favorite book on multivariate stats is the ancient T.W. Anderson book called something like Multivariate Statistics. Of course, TW had no idea what a copula was (probably) and the book has nothing to do with finance. It's very beautiful book though.