arby347 87 posts msg #65650 - Ignore arby347 |
7/31/2008 6:37:43 PM
RIGGS, to change the subject a little:
The Linear Regression (20) indicator observations you made coincides within 3-4 bars from many other things people are posting on these threads. For example Mary4 Moneys post regarding the expected reversals of qqqq and qid.
Have you written any filters to use the indicator? Thx
|
chetron 2,817 posts msg #65652 - Ignore chetron modified |
7/31/2008 7:09:56 PM
the debugged version.....
added a forgotten "williams"
removed a couple "line feed" characters. be sure not to hit the "return" key in the middle of a condition line.
added a few "/*" that were missing.
this removed 66 percent of the results i got. is this still valid, genius?
the debug tab is awesome. great enhancement, SF!!!!!!!!!!
|
WALLSTREETGENIUS 983 posts msg #65653 - Ignore WALLSTREETGENIUS |
7/31/2008 7:21:26 PM
arby -
Writing a filter for LRI(20) would be a waste of time for me because by the time LRI(20) turns up, I'm already in a trade with my "Bottom" filters. I use LRI(20) to look for it's touching or cutting through candles. See, when LRI(2) catches up to a candle and makes contact with it, price begins to either move sideways, or slow it's momentum.
If you're in a serious downtrend, and LRI(20) slices through and gets under a stock, more times than not, I have already hit the trigger on that stock because of it's "touching or slicing through", or because of a combination of other indicators combined. But with that said...it all begins & ends with LRI!
- RIGGS -
|
WALLSTREETGENIUS 983 posts msg #65654 - Ignore WALLSTREETGENIUS modified |
7/31/2008 7:28:23 PM
I should have known better than to ask anyone else other than "chetron!" Excellent work as usual, and you just made "Daves" day! 8^ )
Thanks,
RIGGS
|
dwiggains 447 posts msg #65658 - Ignore dwiggains modified |
7/31/2008 8:10:17 PM
Hi
I surrender. I still can not get this to work. I get 2789 results.
Thanks for all the help
See ya
David
|
chetron 2,817 posts msg #65659 - Ignore chetron |
7/31/2008 8:39:45 PM
wsg,
sorry i didn't catch it quicker.
|
Jay123 9 posts msg #65666 - Ignore Jay123 |
7/31/2008 10:04:55 PM
Can someone put me in the right direction on backtesting this? I'm obviously doing something wrong, as I get a -26% ROI for the period 4/1/08 to 7/24/08.
I removed the 3-day offset, used an 8% stop loss and a 8% profit stop, min holding days = 1, max holding days = 4, maximum 1 trade per day, max open positions = 4.
|
FuriousThug 256 posts msg #65667 - Ignore FuriousThug |
7/31/2008 10:16:38 PM
All of WSG's filters backtest like cr**. ;)
I think they all contain elements of method and interpretation that are beyond the mechanical capacity of the backtesting technology. As a result, I pretty much have to rely on offsetting the filters and just going back and looking at countless examples and their setups to see how they work - that's the only way to do better than -28%, I think.
FT
|
WALLSTREETGENIUS 983 posts msg #65668 - Ignore WALLSTREETGENIUS |
7/31/2008 10:44:20 PM
Thug -
"Backtest?"...are you kidding me? These are designed for 2-3 day pops only! Have you not read ANYTHING about me? Then you go to the next one, and the next....etc. Have you not noticed how volatile this market has been for the last several months?
So tell me...with you being a "greybeard", you must me down well over 20% on the year huh? Why in hell would someone backtest a successful 2-3 day filter? You backtest nerds crack me up! ROFL
- WALLSTREETGENIUS -
|
Jay123 9 posts msg #65669 - Ignore Jay123 |
7/31/2008 10:50:15 PM
WSG... This is a newbie begging for patience :-) Doesn't the backfilter allow an exit trigger that can profitably ride a 2-3 day pop?
|