Kevin_in_GA
4,599 posts
msg #107699
- Ignore Kevin_in_GA |
8/17/2012 1:58:42 PM
Kevin were you able to run the data in back test with the stops as none and then at 1% - 10% yet?
+++++++
I looked at adding trailing stops from 1% to 20%, and just as before the 3% stop is the most profitable. This is one of the few instances I have seen where the addition of a stop (of any kind) resulted in more profit.
I know there are those will will argue, perhaps vehemently, that stops are critical to long-term success. I am not challenging their beliefs. I simply have not seen that many systems that traded better in a rigorous back test with a stop rather than without one.
Without a stop it still did quite well, but had a lower CAGR and a lower Sharpe ratio.
|
Billirider321
62 posts
msg #107799
- Ignore Billirider321 |
8/24/2012 3:53:57 PM
As per the model we should be going long this mornig. Is anybody following this strategy.
thanks
Bill
|
novacane32000
331 posts
msg #107800
- Ignore novacane32000 |
8/24/2012 5:00:18 PM
Bill- The SSO signal to go long was on 8/16/12.
|
duke56468
683 posts
msg #107802
- Ignore duke56468 |
8/24/2012 8:28:17 PM
I was in on the SDS signal but didn't get in until the 15th because I thought it best to wait on some confirmation (not the way it is supposed to be done). On the 16th it flipped to SSO and I should have exited, but it seemed too over sold not to bounce. I sold it this AM for a small gain but was under water 2.5% at one point. I don't fully grasp the logic of the filter, so based on that I will probably paper trade it until my confidence level increases. Usually when I don't stick to the rules I get bit in the rear end.
|
novacane32000
331 posts
msg #107808
- Ignore novacane32000
modified |
8/25/2012 2:39:14 PM
Duke
Kevin can probably explain this better than me but the filter is based on a search Stratasearch did on 1000's of possible STO crossovers and found the one that has worked best for the time period tested.
Now, whether it makes sense technically or not based on what we have learned or been taught on tech analysis I believe is irrelevant because the statistics simply say "this is what has worked best historically".
I absolutely trust this is the best crossover you could use for this particular indicator. Will it work going forward? That is the question.
|
jackmack
334 posts
msg #107817
- Ignore jackmack
modified |
8/27/2012 7:58:40 AM
For chart purposes how would you add the 5 and 10 day EMA?
Thanks
Check that - found in SF 2.0 can just add to charts
|
jackmack
334 posts
msg #107818
- Ignore jackmack |
8/27/2012 8:30:28 AM
bjfeen were you able to update your spreadsheet with the 3% stop loss?
Also - thank you for sharing this with us here - that is some awesome work :-)
Thank you
|
jackmack
334 posts
msg #107830
- Ignore jackmack
modified |
8/28/2012 9:51:36 AM
tconte - just wondering if you wouldn't mind trying the same idea but using the 30 day SMA (vs. 50) as
requested by novacane32000.
Trade SSO signal only above and SDS only when below - and only trade on the "next" signal given
once price moves through the SMA to prevent potential whip-saws and use the 3% stop and do not
take the next trade unless it corresponds with the price trend as indicated by price in reltion to 30 SMA.
Just curious if it improves overall results because viewing a chart of SSO or SDS the 30 day SMA
looks to capture a lot of the signaled trades and by not taking the inverse (counter trend) trades it
could possibly increase the overall equity curve.
Thank you
|
tconte
13 posts
msg #107841
- Ignore tconte |
8/28/2012 11:04:09 PM
jackmack. The result is not impressive. The capital at the end is $313k, and a lot of trades did not qualify, ending up with an average of 1 trade per month. Please see the link below.
https://docs.google.com/spreadsheet/ccc?key=0Ah-vGIsKIZandHlTb3JBZ2Z3UVVaVTd4OGV3XzIwYVE
|
jackmack
334 posts
msg #107848
- Ignore jackmack
modified |
8/29/2012 8:07:28 AM
tconte
Thank you for taking the time to do this - it appears the 50 day SMA is indeed the better approach if one is using this other than the way Kevin originally set out.
Thank you again for your time and efforts.
Do you think this set up with the 50 (SMA) could be scripted so SSO entered on the NEXT signal when price is above or slopping up -- and SDS when the NEXT signal when price is below or slopping down?
|