http://www.multicharts.com/market-data-feeds/ see link, you can get tradestation data into multicharts. I would also be interested if you can do a comparison.
OK. I'll post my findings when I get curious about MC. This will be after I have a lot of fun with Maestro, of course. So many money management tactics are left to be simulated, I see a good few months of development ahead of me. Some goodies I saw so far were equity curve trailing stop, various money management rules (I have been using fixed fractional up to this point), etc. etc. This is the keystone that ties it all together. The system shall work for me....... skynet on a leash!
tailz,i cant program my phone,but i can watch the eur vs trans(not manipulated)vs es or spx,toss in spy and oex and uvol vs dvol for confirmation,is there a way to program that ,its marginally easy to do it in a discretionary mode
If you look at indicators when you make discretionary trades, yes it is possible. Even if you don't, it may still be possible. I learned to trade on indicators so it was a natural progression for me. It's very cool to check the console at the end of the day and see what the robots did. One system has been long M for the past week, the BZH system just went long, GLD got stopped out yesterday because of the overnight move, and the rest of them are sitting it out, waiting for action. The idea is to put a bigscreen LCD on the wall of my shop with the console info windowed all around, so I can monitor the connection while I do something a little more intellectually stimulating during the day. Nothing against daytrading, but you do spend a lot of time waiting. I'd rather be building.
Greetingz, system enthusiasts. Why run a portfolio of systems? 1. Lower risk. If one strategy blows up, your other 99 probably didn't. 2. Volatility reduction. The individual strategies may make you want to vomit, but taken as a group you get a much smoother ride. 3. Position thresholds. Some stocks have a threshold where you exceed the book bid when you take a position. BZH seems to have about $50-100k during the day on the MM's book. I'm a small trader and would rather put $10k on 100 stocks rather than $1m on a single name. 4. All this is obvious to anyone managing a lot of money. 5. But for the amateurs out there (myself being nothing more than an engineer with a trade console, a couple years discretionary experience, and a small library of system trading books), here is another example. Following the theme of this thread: Latest specimen: LCC (US Airways group)
..... And here is how it changes the portfolio equity curve. The trick is to find systems that have different enough characteristics to provide a bodacious final product. Nature follows the same principle: incestuous combinations lead to mediocre outcomes.
2. Volatility reduction. The individual strategies may make you want to vomit, but taken as a group you get a much smoother ride. +++++++++++++++++++++++++++++++++++++ This is my current quest for the holy grail. It's quite obvious that there are market conditions during which my systems don't do well. I need to expand into uncorrelated markets and get more strats so my volatility is like a groomed ski slope. I must say I'm loving running on auto. No more stress trying to decide when to buy and sell. Just let the system do it's thing. If something looks bad, make a change and test it out. I'm still too emotionally invested with each individual trade but at least there is no more second guessing every trade. My stomach feels so much better. I want to be the maestro of the symphony of my bots. The money, like notes, will take me somewhere else. There's so much to do, learn and try. Good thing it's fun huh?
That's the idea. Latest addition to the gene pool: LDK solar. By the way, all I'm doing is looking for a stock with smooth looking price action and placing my existing strategy over it, fiddling with the parameters for about a half hour, then dumping it into the WFO engine. I have forward tested this general strategy on SPY and got usable results for the 6 month test period (training data = 4 years).
The portfolio sharpe is now above .3 and drawdowns are reduced by a factor of 4: For reference, each strategy sharpe has been coming in at about .1 with 20-30% DD's. Maestro is the final step for making hardcore trend following bots palatable for human consumption. And I still haven't gotten beyond stupidly crude money management. That is the ultimate tool for risk reduction and outsized return$. Right Bill?
Wide, do you have total faith that this will work going forward? I ask this because my strats look good so far in their early development but part of me thinks the whole thing is too good to be true. Are you far enough along to have trust in your systems or are you still in the theoretical stage? I will eventually look at PM. Right now I'm still trying to create more strats. But maybe I don't need more strats but just more vehicles.