Let's discuss academic research on mean-reverting trading strategies...

Discussion in 'Strategy Building' started by mizhael, May 24, 2009.

  1. Nattdog

    Nattdog

    The trouble with academic research on strategies is that they are motivated to appear impressive to other professors rather than to generate impressive results. There is a world of difference. Professors are rewarded for status of process/idea and not entrepreneurial results, and for that reason most all of their research (all that I have seen) is worthless.
     
    #41     Jun 12, 2009
  2. because if you can value things correctly, there actually is an arb.
     
    #42     Jun 12, 2009
  3. what's Grey Swan? is it something that happens 1-4 times/year?
     
    #43     Jun 12, 2009
  4. paper trade, visualization-based. no make look at fundamental or news

    detected spike in novartis on the finviz market map 'radar screen'

    NVS -239@$42.002 (total position $10K) Novartis, Switzerland

    EWL +542@$18.46 (total position $10K) Switzerland ETF

    rationale:

    4% spike in novartis (daily standard deviation about 1.3%)
    high correlation ~87% between novartis and ewl
    short major component stock against country etf
     
    #44     Jun 12, 2009
  5. "Novartis can outperform the local index longer than you can stay solvent"
     
    #45     Jun 12, 2009
  6. That's the argument against the trade. In fact, that's exactly why it's a paper trade.

    The argument in favor of the trade is:

    A finance professor and his graduate student are walking along the street when the student spots a $100 bill and stops to pick it up. The finance professor says, "Don't bother to bend down to pick it up; if it were really a $100 bill, it wouldn't be there."

    I'll close the position by the end of next week, and we'll see. I think it's a high-probability trade because the spike is so strong vs the daily volatility.
     
    #46     Jun 12, 2009
  7. moo

    moo

    Haha, good question!

    Perhaps because the quants want to believe there is an arb.
     
    #47     Jun 12, 2009
  8. you're being ironic, but unfortunately you are right.

    i tried to do a visualization-based trade 'long DIA/short BAC' based on the huge spike in BAC

    (although i have to say: even though it looked huge, it was more or less within 1 daily standard deviation based on current volatility, my "rule" is take the trade if it's 1.5-2 standard deviations move or more, like with novartis)

    the BAC paper trade is going against me, BAC has made a 4-5% spike per day, for two consecutive days. will it revert to some mean value vis-a-vis the dow jones? i don't know. it's stupid to be stubborn in this game, and if it were a real trade, i would've closed the trade for a loss of $300-500 (each of the positions is worth $10K).

    so again you need more than one tool in the tool box.

    i don't really know what to think of this DIA/BAC trade. maybe don't touch fragile sectors and companies, where there's too much uncertainty - financials, real estate.

    i mean imagine this trade: go long IYR (the real estate index) and short VNO (one of its components), or another stock like simon property group. well, it may look like a clever trade, but you are going long iyr, and, for my money, that 'leg' of the spread is risky. iyr is very fragile.

    plus you need more than one standard deviation move. and you need to look at the ratio and its patterns, i haven't done that because i don't have the software.

    in any event, all this means all this is not bullet-proof, unfortunately. that's why emotional discipline and conquering your past fears is so important in all this.

    Varima-Garch
     
    #48     Jun 13, 2009

  9. I know from experience. I warned some people but they did not listen. Now they cry. The whole thing is a huge fallacy. Cointegration implies corellation. You know what that means. It means that corellation does not imply cointegration, yes we know that by now. But corellation will be there if cointegration is present. For how long? For as long as it usually takes to take the money from the many and transfer it to the few and this will be true as with any type of trading while your exposure is double in pairs.
     
    #49     Jun 13, 2009
  10. well said! i think the aim is then to try to become one of those few, and take from the many

    but again, it can only happen if you conquer your past fears and unleash inner trading energy. the mouse is your sword and you shouldn't feel subdued by the person on the other side of the trade. it's about domination and control.

    that's how i see it, be it pairs trading or anything else. psychology is key. confidence comes from strengh, and strength comes from weakness. that's the ying/yang alchemical transformation process of trading. very roughly speaking, we need to find a way to transform our losers into winners, and our weakness into strength.

    again, it's my personal opinion
     
    #50     Jun 13, 2009