| Index | Recent Threads | Who's Online | User List | Search |
|
|
 |
Portfolio123 » List all forums » Forum: Simulations and Portfolios » Thread: Fading a system |
|
Total posts in this thread: 4 |
[Request a Feature]
[Post new Thread] [Add To My Favorites] [Watch this Thread] |
| Author |
|
|
BJS
Advanced Member
|
Here's a portfolio I've had for over a year now. It has not been breaking any performance records, but it has been doing something interesting. Look at the performance chart, and look what kind of a support line you could draw under it - almost a perfect straight line. http://www.portfolio123.com/port_summary.jsp?portid=90050 So, with a system like this, how about fading it? In other words, let it do its own thing, but only buy the stocks when the equity line drops down to the support line. Then when it runs up, sell the stocks and wait for it to drop down to support again. Any thoughts? I haven't tried it yet, but have been thinking about it for awhile. Brian |
||
|
|
p123robert
Advanced Member CANADA Joined: Nov 7, 2005 Posts: 372 Status: Offline |
This is a great idea because, this way, you could ride a port pretty much only when it's really-really rising, and you could make tens of thousands of dollars in a course of only a few days. As long as you've got highly predictable levels of both support and resistance, this technique is going to work for you all the time! The trick is to have a... 1 ) Highly predictable support level, 2 ) Highly predictable resistance level, and 3 ) As few stocks in the port as possible. There are great many ports that satisfy all these requirements. However, sometimes, even some of the best of ports, for some unexplicable reason, seem to get out of hand. Like this port of yours... if you do a 5-year sim on this port of yours, you will see that, at around January 2006, suddenly and inexplicably, the support level broke, and this port became unreliable. What will you do then? I guess, you want to have emergency plans, and decide in advance, what you will do, if this scenario ever occurs. Then, you could and would cut your losses, and try another port, at that time. Robert |
|||
|
|
faschim
Member
|
In the late ninetees we modelled quite extensively "model timing" ie trading based on the equity curve of a system or model. The results were not encouraging. I would probably use more time finding a system with a more stable result than tweaking a result of tweaking, if you get my drift. (A system/model is by definition tweaking) After all, a model/system is prone to change. An equity timing model based on a model/system is like a second derivative of change in my opinion. |
||
|
|
p123robert
Advanced Member CANADA Joined: Nov 7, 2005 Posts: 372 Status: Offline |
Hi Brian, It's very nice to see you're active again! Here's an idea similar to yours. How about 10 decent ports, with annualized returns in the 50-90% range, where each of them is based on a different ranking system? You create them at the same time, and put all 10 of them on "Auto", and keep an eye on them. How about putting money on the 2nd biggest looser? The worst looser is probably a disaster, but the 2nd biggest looser probably has the greatest upside potential. Then you wait, and spend most of your time running to the bank... to deposit all that cash into your bank account (which should be easy). Then, after a bit of time, once your 2nd biggest looser becomes your 2nd biggest winner, you close all of your positions in that port, and start putting all your money on the 2nd biggest looser port. I hope this helps. Robert ---------------------------------------- [Edit 1 times, last edit by p123robert at Nov 13, 2006 8:33:46 PM] |
|||
|
| [Show Thread Printable Version] [Post new Thread] |
Re: Fading a system
