I have a system, now what

Discussion in 'Professional Trading' started by dude_03, Jan 31, 2013.

  1. dude_03

    dude_03

    I have been working for >2y years as a quant in a prop desk in a bank, while developing a automated system overnight and on the weekend.

    I have been paper trading my system at work for ~8months now, and it proved to be good - where with "good" I mean that even under some conservative assumptions, it proved to have good risk-adjusted performance.
    Great, you would say, where's the problem...the problem is that for for a number of "political" reasons, I am not allowed to make the jump to trading where I am, nor I am given a realistic indication on when this could happen.

    So I am evaluating my options, and would like to receive some advice by some of the very experienced guys populating this forum.

    Of course I am not saying that the system is failure-proof, I haven't even traded it live yet (!), but at the very least is showing some great potential.
    Current idea is to first start trading my own and possibly family money to develop some practicality with operational matters (not much $, but enough to trade with minimum quantities while having the right risk/capital profile).

    The possibilities after that are:
    -open a CTA, and start trading with my money to begin with. If it does well, I think I can find some relatively high-net-worth individuals willing to invest ("find" as in I know them directly or via a second person).

    -join a prop firm. I am a bit concerned in this case on how Intellectual Property would be handled, especially because I don't have an actual track record to make them "trust" me. Also I am not sure on how much capital I might be given (any opinion/experience sharing on IP and capital matter in this case would be greatly appreciated of course...)

    If everything goes very wrong, and in say 1year I have gone nowhere and lost/finished all the money, I hope to be able to find another job- timing for this is not great, but 2.5 years as a quant in a prop desk should be worth something.

    Little background on the system:
    currently intraday only, with 1-2 round trip per day per instrument;
    operates on FX and fixed income futures;
    Backtested on 12 years of data, overall Sharpe 2, win rate >~50%, average win/average loss = 4/3 (including transaction costs, bid/offer, slippage).
    I believe these factors could make it quite marketable as different from most of other funds/CTAs...opinions pro/cons are more than welcome;

    Personal Considerations:
    Of course to do any of the above, I would have to quit my job, giving up the option of getting to trade in there.
    I am on my late 20s and don't have family/mortgage to sustain yet, which helps in making the time for such a move right.
    While I am kind of learning about markets where I am, the "theta" is quite high: the desk doesn't do much algo stuff, so I am learning about different styles (which I am not intended to pursue) and the idea that I am wasting my time (which I consider one of the most precious things one has) doing something that I don't like irritates me as hell.


    Why I want to do it:
    Interestingly, it's not for the money itself. A quant career can be very profitable and almost riskless/stress free, and mine is well launched at the moment. If we count the number of people who "made it" trading, I think odds say that I should stick to quant.
    However, for me it's about doing something that I enjoy doing. I would be way happier to trade and earn less than following a different –richer- path.

    Thanks in advance for any advice/suggestion/opinion/insult for being nuts
     
  2. Is there any way you can do both?
     
  3. Is your dad retired by any chance? If the system is automated, maybe you can teach him or whoever you trust just enough to monitor the system throughout the day. They would just need to know how to flatten all positions and turn it off in case something went wrong right?

    There could be a ton more to it than that, I am just trying to think maybe a route like that could make it feasible. Then if it works well you can leave your job as a quant and continue running the system on your own.
     
  4. oraclewizard77

    oraclewizard77 Moderator

    Think of trading as a business. You are the owner / manager.

    Since you currently have a job and are not sure moving a paper trading system to real money is feasible, you have some options.

    1) Hire someone to monitor the system for example a trader from this website who has the time to trade and watch your system. Part of the purpose is to also shut the system down if it starts to go crazy like that updated automation from Knight which almost destroyed their company.
    a) You could pay him or her an hourly salary.
    b) If the system is really good, pay him or her with some profits of the system.
    c) A combination of the above.

    2) Quit your job and trade your system yourself.

    3) Sell your system to a hedge fund.

    4) As you stated before, become a CTA and raise other investors money.
     
  5. CT10Gov

    CT10Gov

    Quant trading is an actual thing. Just because your prop group doesn't do it, why not go look for another shop that does, be it sell side or buy side.
     
  6. How many quants lost their pants in the history?it was a rethorical question.
     
  7. CT10Gov

    CT10Gov

    Funny coming from a guy looking for magic numbers a few days ago.
     
  8. What panty less quants have to do with the magic numbers?It was a rethorical question again.
     
  9. gmst

    gmst

    12 yrs backtest is good. Your training as a quant and experience is useful. But you never traded - that is bad.

    Fee free to papertrade and analyse for years but in reality there are just 2 ways to actually do it:

    A) My Way or Highway - In this case, you go to your boss and declare you are not coming from tomorrow. You are going to trade your PA. You decide RESOLUTELY that you are ending your banking career and you are going to give yourself 5 years to succeed. Btw, 1 year will just not cut it. LOL. Everyone thinks in the beginning 1 yr will be sufficient to make it in trading. I actually know a guy who used this approach! Ex-MS and Goldy - but he has still not made it but it has been just 2 yrs.

    Frankly, this is borderline insane way to do it. And the only real reason imo, to go this approach is if you can do 4x or more return per year on your starting capital and have at least 50k as starting capital. So, to gauge this - do this little exercise:

    Tell us how much PERSONAL starting capital you can invest, how much leverage and how much DD you willing to accept if you leave your job and trade this for a living? Now use this number for personal starting capital and personal leverage you are comfortable with, compute the equity curve and I would also say post it here.


    B) A more sane and more risk-adjusted ways to do this:-

    Approach a good buyside firm like Millennium partners with your strategy. I heard they are always hiring on a profit share deal. Only problem you will face is that you do not have a live trackrecord. But if you know people there and can get an interview, I would say it is a good shot. Even if they don't hire you, you will end up learning something from the interview.
     
  10. dude_03

    dude_03

    Thanks everybody for the responses.

    I agree that the bottom line is whether I can be confident enough on the system to actually perform (although even if you have an actual track records you could ask yourself the same question)

    In response to those suggesting to have somebody/my dad monitor the system:

    No, my dad is not retired :)

    But seriously, I see quite a few problems with this:
    1- I can't really hire a monkey: being intraday, the system is not as simple as send orders on open/close, and forget about it.

    2- To have reliable results it would have to be monitored
    for quite some time, say 3m at least I guess...that's quite a bit of money. And it's not as simple as paying somebody with the proceeds: starting capital is small, profits will be small. (say 50k capital, assume a very honest 50% annual, that's only 25k in one year pre tax)

    3- I don't know if I would trust somebody with details of the system - and if I don't share them...well, that's almost the same as running it on auto-mode.




    I don't want to sell my system to anybody!
    I want to trade it and eventually open my hf/CTA, if I am good enough.
     
    #10     Jan 31, 2013