calhawk01
135 posts
msg #102182
- Ignore calhawk01 |
8/11/2011 11:29:49 AM
Anyone know how I can create a filter that I can add to "indicator" section of backtest that gives me the ENTRY DATE low, high and close etc
Thank you
|
calhawk01
135 posts
msg #102214
- Ignore calhawk01 |
8/12/2011 4:37:01 PM
Anyone? I need to get a filter that I can add to "add indicators on backtest" for the day after the trigger date. Please help.
|
four
5,087 posts
msg #102225
- Ignore four
modified |
8/12/2011 10:18:20 PM
ENTRY DATE low, high and close
-------------
After clicking INDICATORS, enter the following 3 lines into the area 1. ENTRY INDICATORS:
add column low
add column high
add column close
|
four
5,087 posts
msg #102227
- Ignore four
modified |
8/12/2011 10:43:03 PM
I need to get a filter that I can add to "add indicators on backtest" for the day after the trigger date
-----------------
perhaps a work around...
I suppose you could (1) use the DOWNLOAD CSV feature to Excel (2) Add commas between the symbols (3) Copy the symbol list into a filter (4) add line offset mm/dd/yyyy (where the day is the day after the trigger date of the backtest).
|
calhawk01
135 posts
msg #102241
- Ignore calhawk01 |
8/13/2011 4:16:21 PM
I mean if the holding period is 0 then the drawdown and max gain can also give us low and high theoretically but the whole point of doing this is to verify validity of SFs backtested results and data. If you go to backtested part of these forums, you will see one of my post about it.
I wish there was a way to do t+ trigger date that we can enter in indicator. I see what you are saying using offset date and symbols in a filter. It wouldnt work since I have 50000 entries. Find their entry date alone can take a while and then offsetting that date.
And the add column low, high only gives us the trigger date statistics.
There's got to be a way!
|
four
5,087 posts
msg #102245
- Ignore four
modified |
8/13/2011 8:01:23 PM
calhawk01
Send a request directly to Stockfetcher. See what happens...
|
calhawk01
135 posts
msg #102249
- Ignore calhawk01 |
8/14/2011 1:35:05 AM
I actually did, the guy stopped responding to me. It's rediculas. Their data are not worth anything. Stockfetcher is great for developing ideas but their backtested data is HORRIBLE to say the least. I kid you not, i downloaded this excel file for one of my backtested results a week ago. That file according to backtested results had 87% winners. So i did a little IF statement in excel to verify that. The data showed that indeed it had 87% winners. Two days after, i downloaded the same data again from stockfetcher because i was at work. I did the IF statement again on the same data, it only had 57% winners! Yet on SF website --> backtest __ still showed 87% winners. Something is for sure fishy with the data here. And that is why i'm trying to check all of the data. I have literally spent months working on this one strategy. And all of a sudden i came to the realization that all the data is worth nothing. I'm kinda pissed off. And would like some answers. I mean aside from the winners, another thing in the file is that -- > the Gain/loss section. Let's say that my holding period is 0 days. My profit target is X%. And my profit target is HIT according to SF. Then, why in the world would SF record a gain/loss that is MORE than my profit target??? This changes EVERYTHING. This changes all of my results for ROI. This is so frustrating. SF has so many features that are good, but if their data has over 20% error in it, all their results are worth nothing.
|
calhawk01
135 posts
msg #102250
- Ignore calhawk01 |
8/14/2011 2:12:35 AM
Look at this;
DOW 1/15/2002 ENTRY: 25.6 1/15/2002 EXIT:25.86
25.6 is my entry according to SF. According to YAHOO HISTORICAL PRICES:
Jan 15, 2002
OPEN: 27.25
HIGH: 27.74
LOW: 26.36
CLOSE: 26.45
THE PRICE WAS NEVER EVEN CLOSE TO THE REALIZED ENTRY THAT SF DID FOR MY BACKTEST RESULTS! AND THE EXIT! And thats a X% gain added to my "ROI." This is useless! I have 50000 data points, and I just found that, but randomly picking a trade. There must be thousands of bad trades like this.......................................................................
|
calhawk01
135 posts
msg #102256
- Ignore calhawk01 |
8/14/2011 2:10:12 PM
And you just have to LOVE the response from this community regarding this. THIS MEAN THAT ALL YOUR 10000000000000%%%%% ROI giver filters with 100%%%% WIN RATIO, ARE NOT RAN AGAINST CREDIBLE DATA................ YOU guys maybe just maybe want to check your DATA
|
duke56468
683 posts
msg #102257
- Ignore duke56468 |
8/14/2011 2:16:07 PM
Is there a problem with other holding periods also or is it just with the 0 holding period?
|