From your reply I can tell you are a seasoned trader. I like the content of your response. It states much better what I was thinking when I wrote my response. Thank you.
estim, your posts sound like you are an academic person. But strangely, you seem to believe that its possible to achieve success here without some original idea - I assure you that in the real world it is just as crucial to have new ideas as in the academia. Similarly, new and bright ideas are worthless if it isnt accompanied by hard work (of course), but hard work alone does not get you very far in quantitative trading, just as in academia. Just in case someone accuses me of elitism, let me clarify, I dont believe that academics have monopoly on ideas. It is pretty clear that even uneducated people have original, fresh, practical, ideas. A good example is McDonalds which is a huge corporation built on top of a single, very valuable idea. Hell, even big time mafiosi had built their criminal empires on some idea. I am sure everyone knows plenty of other examples, but the real point is that <b>every</b> big business success is due to some good idea. Of course, this wont help you in the constructive sense, it is not possible to sit down and decide to have ideas - one either has them or not.
ok, since my first cruel message, i thought of a little bit more positive suggestions. I often hear people complain that they have tried a few things and none of them "worked". But what exactly do you mean by that? Let me guess... for most of my acquaitances, it meant: did not find a magical recipe to make 100% per year preferably without risk. Less than 100% per year, or maybe per month, was unacceptable because they only have 10K or 5K capital, and any less return is not worth their time. If that is the case, then of course, it is indeed not worth your time, and it never will be. But if its not magical recipe you are looking for, what does it mean you didnt find "anything"? Suppose you are looking at some market. Absolutely any strategy will either make money or lose money - after you subtract the return from buy&hold. If its negative, you can just invert the rule, and make the return positive. That is actually not always true, because if you are already modeling the return taking into account the transaction costs, or bid/ask spread or whatever, then the return may be negative even with the opposite strategy. (if that is the case, i sugest to think of your return as positive regardless, but the transaction cost as the minimum threshold to be overcome as you improve) OK, so what you are left with, in the worst case scenario is a very low Sharpe ratio, or information ratio, or whatever other criterion you are using. But it is not NOTHING, you just have to improve on it over time.
kotika, I read your first post, walk around in the living room and thought it was amazing how many that actually would spend some time on useless comments here on ET., especially people who are not doing fully automated trading. Then I came back to the computer and read your second post that had a little more positive attitude. So here are my comments. Yes you are right about the big magic system, for sure I have had something which I have considered not to be enough. That is why I am going back to some of my old setups which I will run in parallel depending on market conditions and larger time frames. Since I live in Denmark the time for trading us stocks is from 15:30-22:00 and I am not interesting in sitting glued to my monitors during those hours, this is why my focus is on fully automated systems which of course need adjustments and monitoring but not as intense as discretionary trading. I am not looking for the big new idea, which I donât get because my economic or discretionary trading knowledge limits me. I have changed my original goal of making huge systems to make a system that will give me an additional income. estim
I have to disagree with kotika, who says you need an original idea. No need to reinvent the wheel. there are plenty of useful ideas out there. heck there are plenty of useful ideas amongdt all this crap at ET. you just gotta find em.
with so much noise out there now to capture . a pairs strategy with a longer time frame is perfect. Trading a relationship between 2 equities, you need time to capture a reversion. A short term strategy does not give you enough time to capture a big move and reversion. I never thought I'd see the vix above 30, we have been above 40 for the last 5 months take advantage....
Think about it, who are you competing with in those time frames? Analyze carefully what kind of edge it would need to compete with these multi-billion Dollar quant HFs & desks on a leverage, trading cost, redundancy, exchange access & speed basis. Try changing your time frame to EOD or weekly or try changing to less liquid small/micro cap stocks -- where there's less sophisticated competition.
As far as i have heard, nobody turns on their black box and goes to have fun. At the very least, the original developer hires someone to "babysit" the computer, just in case one of a million things can go wrong. This person should be authorized to press the "CLOSE ALL POSITIONS" button (if anything at all is suspicious). But thats if you have an automated trading system, which is a hard thing to achieve - i dont even have it working yet, after thinking about it and making small steps for a very long time. For the first three years, i had only quant models, where i used the computer to guide me, i make the final decision myself. I was quiet successful, until recently. Now, I am starting to work with a trading API and as you know the computer supposed to make the trades. I am finding it is very tricky to get all the implementation details right, even if your basic strategy works well. I have something which is very much based on my previous models, but now adapted to shorter timeframes. Nice bars around Strøget. I wish i was young again...