|
|
Rank: Newbie
Groups: Member
Joined: 7/3/2010
Posts: 4
Points: 12
Location: USA
|
Hello,
i'm new here and very impressed with the strategy. I've read thru most of the forum posts and blog entries to get up to speed.
One question is about using trading stops. Has anyone backtested the strategy with stops? I see these results were done without stops:
http://blog.mtrig.com/mtrig/blog/page/-Trading-Strategy-.aspx
If someone had backtested, what were the results?
Thanks!
|
|
|
|
|
|
Rank: Admin
Groups: Admin
Joined: 4/25/2009
Posts: 74
Points: 200
Location: USA
|
Hi BD,
I back-tested the system with stops previously, the one issue I came up with was... when to get back in (for example long) if a stop was hit but the signal did not change.
I can post the results to this thread with a 5% stop loss. If you want to see something more such as 8% let me know.
I can have the results posted some time this weekend.
Thanks,
Patrick
|
|
Rank: Newbie
Groups: Member
Joined: 7/3/2010
Posts: 4
Points: 12
Location: USA
|
Patrick,
Yes, I was wondering about the trade-off of stops with whipsaws. I backtested IWR from 10/3/01 to current and got about 15% annualized return which is certainly better than the overall market. But as you would know, 2010 was down because of the recent volatility. I would like to see any additional backtesting with stops that you have time to do. Thanks!!!
Bryan
|
|
Rank: Admin
Groups: Admin
Joined: 4/25/2009
Posts: 74
Points: 200
Location: USA
|
Bryan,
The figures below are from 1/1/1997 to 7/9/2010
I. Holding from Signal To Signal - No Stops
II. 5% Trailing and 5% MAX Stop Loss
Thanks,
Patrick
|
|
Rank: Newbie
Groups: Member
Joined: 7/3/2010
Posts: 4
Points: 12
Location: USA
|
Well it is clear that the stops prevent you from riding out the longer, positive trades. This does makes sense, especially in our more volatile market.
Patrick, I really appreciate that you did this.
Did you run this backtest on IWR or one of the other mid-cap ETFs?
Thanks again,
|
|
Rank: Admin
Groups: Admin
Joined: 4/25/2009
Posts: 74
Points: 200
Location: USA
|
I ran it strictly on the Value line symbol (Yahoo! ^VAY).
On another note, the model had some serious whip-saws. This is by far the worst since I have been using the timing model. Based on the logic I do not see a way around it at this point. I will have to peer into it again to see if it can be improved further.
|
|
Rank: Newbie
Groups: Member
Joined: 7/3/2010
Posts: 4
Points: 12
Location: USA
|
I know from reading your forum and blog posts that you have done a great deal of testing and optimization. When I first found this site, I did a backtest and mistakenly used prices from the day of the signal (vs the next day) to test. The results were more than twice as good, over 30% annualized return. Of course, this is incorrect and if the signal criteria was changed/relaxed it would generate more signals which would likely reduce the results. But this, along with your stop analysis shows the power of riding the trend, especially if you get in right at the beginning.
But I do wonder if there is a way to use your 'early warning' indicator to trade? I like how you added this feature, have you tested any trading off of this?
|
|
|
Guest |