Kevin_in_GA
4,599 posts
msg #108009
- Ignore Kevin_in_GA |
9/11/2012 11:05:37 AM
Highest APR, low drawdown, high Sharpe ratio. Those were the criteria I loked for.
This specific combination was the highest APR and among the highest Sharpe and lowest drawdown.
|
duke56468
683 posts
msg #108011
- Ignore duke56468 |
9/11/2012 1:15:26 PM
On this last trade a trailing 3% stop would have saved you several hundred dollars (depending on the size of the position) over a hard 3% stop. However in the backtests I did, a trailing stop did considerably worse over a 4 mo period.
|
Kevin_in_GA
4,599 posts
msg #108012
- Ignore Kevin_in_GA |
9/11/2012 2:38:06 PM
Note: The addition of a 3% trailing stop here generated the best overall performance since 1/2/2007, as I had shown in the first post. The last four months is not necessarily representative of the system's long term performance.
|
jackmack
334 posts
msg #108049
- Ignore jackmack
modified |
9/14/2012 8:44:35 AM
Just wondering - is there any statistical data that could be added to this that would indicate if a "stop" is hit in one day to take the opposite trade and just set your new stop at the buy price of the new position - this most recent SDS stop switching to SSO once the stop was hit. Again this would be for a one day stop being hit (a gap opening up or down).
Again - IF the stop occurs in ONE trading day - not a gradual day over day hit.
|
Kevin_in_GA
4,599 posts
msg #108050
- Ignore Kevin_in_GA |
9/14/2012 9:32:28 AM
Not that I am aware of - that type of analysis would probably need to be done manually and I am not sure it makes sense given the overall historical results just keeping the stop as is and re-entering at the next crossover.
|
jackmack
334 posts
msg #108051
- Ignore jackmack
modified |
9/14/2012 9:52:18 AM
Okay
I can understand that. Just looking at the times in data where stops were hit in single day the rally continued several days or more in the new direction and appears that the payout might have added some nice gains and to boot the next signal usually is to have switched anyway so....
I still think this filter rocks and can not thank you enough Kevin for sharing.
Cheers
Edit - this is only from my review when using this filter and the 50 SMA as novacane32000 and tconte worked on previously on pg. 7 of this thread.
|
coolsf
21 posts
msg #108065
- Ignore coolsf |
9/16/2012 2:04:07 PM
So, is the signal now to go long SDS?
|
novacane32000
331 posts
msg #108066
- Ignore novacane32000 |
9/16/2012 2:14:36 PM
cool-the system went long SDS on 8/31/12 but was then stopped out using a 3% trailing stop.
Open the filter-click on SDS and you should be able to see this on the plot near the bottom.
|
coolsf
21 posts
msg #108067
- Ignore coolsf |
9/16/2012 2:20:20 PM
novacane, Thx. It helps.
|
Kevin_in_GA
4,599 posts
msg #108085
- Ignore Kevin_in_GA |
9/18/2012 11:29:27 AM
Intraday today the signal is to go long SSO.
|