[amibroker] Re: ATR./Turtle Script

2010-04-13 Thread whitneybroach
What code change would cause this to exit a position intraday at a 30-bar low? --- In amibroker@yahoogroups.com, three_percent se...@... wrote: Robert, I haven't tried to optimize for CAR or RAR. I usually go for CAR/MDD instead when I run tests. Andy --- In

[amibroker] Re: ATR./Turtle Script

2010-04-09 Thread three_percent
Robert, I haven't tried to optimize for CAR or RAR. I usually go for CAR/MDD instead when I run tests. Andy --- In amibroker@yahoogroups.com, Robert Chevallier robert.chevall...@... wrote: Hi Three_percent What are your settings to get 100% CAR or RAR on SP500 ? I'm far from it

Re: [amibroker] Re: ATR./Turtle Script

2010-04-08 Thread Robert Chevallier
Hi Three_percent What are your settings to get 100% CAR or RAR on SP500 ? I'm far from it trying your .afl script 2010/3/30 three_percent se...@md.metrocast.net // ATR/Turtle Script by Andrew Senft // Thank you very much for your notes. I took out the ExRem statements out and now get

[amibroker] Re: ATR./Turtle Script

2010-04-05 Thread three_percent
Hi Dubi1974, I do have a program that does it percentage wise rather than point wise. For some reason, it works a lot better point wise. You are right though, it just won't trade certain stocks. Andy --- In amibroker@yahoogroups.com, dubi1974 gonzale...@... wrote: Hi three_percent,

[amibroker] Re: ATR./Turtle Script

2010-04-03 Thread dubi1974
Hi three_percent, looked at your code. The performance for stocks seems to be nice. Just do not understand, why you use e.g. for BuyThreshold a fixed value (like 0.3). If you test it on SP stocks like Google, which have always higher ATR than 0.3, they would never been traded. You could try to

[amibroker] Re: ATR./Turtle Script

2010-03-30 Thread three_percent
It' a trend channel finder that computes a tight channel by way of ATR. The lower the ATR, the tighter the channel. It checks for a trend that keeps on going up. Notice the half the ATR period checks. Also, it gets out if the ATR gets too big or it makes 25% for the trade. I'm just

[amibroker] Re: ATR./Turtle Script

2010-03-30 Thread WiseStockTrader
Hi three_percent do you mind if i put this formula on my site for others i will leave all credit to you of course? --- In amibroker@yahoogroups.com, three_percent se...@... wrote: It' a trend channel finder that computes a tight channel by way of ATR. The lower the ATR, the tighter the

[amibroker] Re: ATR./Turtle Script

2010-03-30 Thread three_percent
Wisestocktrader, be my guest and post it on your site... Andy --- In amibroker@yahoogroups.com, WiseStockTrader wisestocktra...@... wrote: Hi three_percent do you mind if i put this formula on my site for others i will leave all credit to you of course? --- In amibroker@yahoogroups.com,

[amibroker] Re: ATR./Turtle Script

2010-03-29 Thread three_percent
Missing a semicolon after 3000 --- In amibroker@yahoogroups.com, three_percent se...@... wrote: Here is my attempt at a turtle script based on the ATR command. I've been backtesting it against the SP500 since 1990. If you guys have any comments or improvements, let me know...

[amibroker] Re: ATR./Turtle Script

2010-03-29 Thread Mike
Hi, Without evaluating the actual logic of what you're trying to do, a couple of quick observations: 1. I suspect that you're missing AND after 3000, not semi colon. 2. You are redundantly recalculating several arrays, where single calculations stored in variables would result in faster

[amibroker] Re: ATR./Turtle Script

2010-03-29 Thread three_percent
// ATR/Turtle Script by Andrew Senft // Thank you very much for your notes. I took out the ExRem statements out and now get 100% annual returns on the SP500. A lot more than before. Not sure if that is right. Also, I'm not sure how to get the initial occurance of the buy trigger for the

[amibroker] Re: ATR./Turtle Script

2010-03-29 Thread Mike
You're still redundantly calculating arrays (e.g. Ref(Close,-1 * ATRPeriod / 2), Ref(Close,-1 * ATRPeriod). For the rest, it would help if you gave a non coded (e.g. english) description of your rules. As for performance, it's good practice run the backtest in full details mode (see AA