Non-Linear Cointegration Trading Model

Discussion in 'Strategy Building' started by FAROOQAMS, Jul 9, 2005.




    Any one have any idea on Cointegration trading model either Linear or Non-linear, How to implememint on excel .....:confused:
  2. ummm..

    I got to translate this into "Plain Speak"

    You want a spreadsheet that intermingles groups of uncorrelated instruments into a line or a jumbling of data?

    Michael B.

    P.S. I obviosly am in the wrong forum (Strategy Trading), I never know what these guys are talking about in here...

  3. I love plenty of these kind of guys, as long as they pump some money in the market.
  4. You know, tons of money is made with cointegration and related concepts. Laugh it up. :D

  5. Yeah, u must be talkin about n*tr*.
    AFAIK, he is in mainly the non-linear kind.
  6. man


    we tried to use cointegration for pairs trading. never used it. other things did better.

    cointegration is just another way of analysing joint movements. in this respect it is very similar to correlation. nevertheless a juri MA is a moving average and you might find something using it which you did not find using a simple ma. so cointegration might as well pay off for you.

    i learned that sophisticated sounding technology means zilch. there are guys who use very simple technology and they build trading systems you can only dream of.

    problem with tougher technology is that it takes more implementing time before you find out that you wasted your resources ...
  7. very sound observation!
  8. man


    and as so many others i could not believe it before i did it on my own. man, if i recall all that sophisticated sounding stuff ...

    BUT there is one great advantage of more complex strategies: the entrance barriers are much higher, just because you need a decent operation to let someone do something for a year without knowing whether it will pay off or not.
  9. pjbreen


    I ran across this paper a while ago. I can't make heads or tails of it, but maybe it will help.

    There's also some free econometric software here that may help you along.

    good luck.

    I just find it interesting the number of VERY high paying quant and hedge fund jobs that require experience w/ cointegration. Makes you go hmmm.....
  10. I agree with this in general, trying something and realizing it doesn't work does have some value. But a *year*?! It only took me about a day or two playing around with cointegration models in matlab to realize it was well worth it.

    Correlation and cointegration are related but very different. Correlation is like, two people walking down the street side by side. Cointegration is like, two people walking, one person stops to tie his shoe, runs up, has to wait on the redlight, and then sprints to catch up again.

    #10     Jan 11, 2006