Chameleontrader - Good question. Being new to this, I wanted to see if papertrading with live data was different from out-of-sample back testing. Looks like maybe the differences are small, if any. But also, my best back tests use tick data. TS makes limited tick data available. So, to do more out-of-sample testing, I'm using live tick data. Thanks, Norm
Hey Norm: Look at these two quotes from you: "It's an intraday trading program (it typically trades 10 to 20 times per day) and I need more sideways and slightly down days." "The system is design for use in an IRA. That is whaere I presently do my trading. So, at the moment, it only goes long." If you cannot see why an intraday trading system which only goes long and needs slightly sideways to down movement will lose money, then you need to do more testing.
Hello Optioncoach, I need more sideways and slightly down days so that I can see, real time, how the system performs under those conditions. I can do this by looking over my backtest results. However, I just wanted to watch some different market conditions live. These are not the days that I expect the most profit. Although, I do expect the system to be profitable on many of these days. It seems to be only the days having big dives that are unprofitable. Thanks, Norm
Why don't you try live-trading with archived data. "Looks like maybe the differences are small, if any."
I have tested with one minute bar archived data extensively. It's okay, but not the same as tick. I am trying to find more tick data. I tried tickplusdata.com. Their pricing is good. But, it looks like you need third party software to load their data into Tradestation's data server. Seems like someone should offer tick data that is easier to use with TradeStation. Any suggestions. Thanks, Norm
MAESTRO you may be a lot of other things, but you are NOT funny. and if you stretched your brain surgery idea to infinity to prove the superiority of your own, powerful set up, you won't get funnier. so sorry.
Second that. Making a joke out of anothers attempt to create a trading system is quite childish, regardless of what you think of their strategy.
this sounds good. the number of trades is high enough to give you quick answers to the extent you might have curve fitted or not. i would reconsider the long only approach. picking shorts is difficult. on a purely technical basis i do not anyone who does it sucessfully, but you might consider shorting a big index against your exposure. but then again, you do not run overnight positions, do you? i do not run anything on single stocks, so i am not the expert. but i would look at my returns in up and down markets on a daily basis. if i was profitable on both (do not know if that is possible long only singles), i would not care at all. if not, hence the long bias in your database and overall stock drift. how long was your backtest? if it did not see a bear market, i'd be scared as hell. if it goes back to 2000 ... test another two weeks, then give it a try. good luck
i don't post here much due to the fact there are more people interested in putting people down while comfortably hidden behind an anonymous id than actually helping them to trade...which is what this site should be for. don't know if this was said yet, as i didnt read all of the posts- but make sure you back test on out of sample data - curve fitting is an easy thing to do as you tweak your program day to day on live data, trying to make it so your program hits that nice trend of the day that it missed before. also take a look at how much of a difference trading out of an IRA will make - commissions, taxes, etc...