Message Font: Serif | Sans-Serif
 
No. of Recommendations: 1
Hi, I'm new to this board so I'll give a little back ground before my question.

I have been working on an automated trading program for the last five months or so (coding in Java, concepts from Covel, Faith and Elder and others). My idea was to apply trend following rules to the Hidden Gems picks. I figured that these stocks due to their volatility and good fundamentals would be more likely to trend than just any old stock. The systems I'm working on currently look for the main trend, wait for a pull back and then enter in the direction of the main trend if it resumes. Surprisingly, in my simulations, I tend to make just as much money on the shorts as I do on the long. Not always, it depends on the particular parameters that I've chosen, but more than I expected (Hidden Gems should always go up, right??). This is actually the reason that I wanted to get into trend following. I've made great returns in HG over the last year and a half, but I want a system that can make money when the market goes down as well as up. I think the dollar is headed down and I'm more than a little worried about what will happen when all the baby boomers start cashing out their stocks to enjoy their golden years (hopefully their $$s will still have some buying power...)

My question is why are futures or ETFs most often associated with trend following systems? Whats wrong with stocks? Are futures more 'trendy'?

Any enlightenment would be greatly appreciated.

Jon

PS. My sims show that if you consider all the HG rec's from before 2005, take only a limited number of positions based on trend efficiency (weeds out the less trendy stocks) you can turn $100'000 in Jan 2005 into about $250'000 in the middle of 2007. Max draw downs are typically about 10-15% and there are minimal margin requirements. The system is set up so that you just have to enter orders and update stops after the market close (< 30 minutes per day). This is pretty much what I have been looking for so I will start trading it in the mid term future if the last of the testing works out (longer term, different group of stocks...)
Print the post Back To Top
No. of Recommendations: 1
Futures are usually associated with trend following for a couple of reasons; trendiness and leverage.

It's pretty clear that stock market indexes are not as trendy as they used to be and there is debate as to whether or not commodities will continue to be. As long as there are humans and the basic laws of supply and demand are not revoked there will always be trends. However, given the sophistication of markets and trading programs now days it is also pretty clear that the signal to noise ratio has increased in just about everything.

Then there is leverage. If you read up on the "great trend followers" you will learn that several employ a basket of leveraged futures to make their money. With a lot of study you can pretty much figure out how much money you need to survive the inevitable drawdowns...and when you are "on" with that much leverage...you are going to make a lot of money. If you've read Covell or Schwagger you will detect that the issue for these guys is often finding the line between making money and scaring away investors due to volatility.

Having said all the above, a single stock is likely to be much more volatile than several commodities and most certainly any stock mkt index.
Print the post Back To Top
No. of Recommendations: 1
Thanks for that info.  It makes sense I guess.

Just to follow up on my initial post, I back tested my system with a different group of stocks back into the early 90's.  The system did great until 2000, and then it got hammered with a 50% draw down that it didn't really recover from until about 2006. Not exactly what I was hoping for.  In fact, with just about every long term run for all the systems I have tried the period between 2000 and 2002 is just brutal.  The question in my mind now is will using the Hiddem Gem picks (small cap, lots of volatilty) help prevent something like that in the future with this system?  I don't really believe so, but there is now way to test it.  In the mean time, I will look for a way to filter out markets that are behaving 'badly' and see if I can make the system somewhat more robust.

Ciao

Jon
Print the post Back To Top
No. of Recommendations: 0
You could use trendiness as a filter. (p213 in Tharp's TYWtFF)
Print the post Back To Top
No. of Recommendations: 0
Jon,

Have you been to the Mechanical Investing Board? If you have not you need to.  Backtested stock systems galore with a superb set of folks who provide all sorts of analysis with regard to them. The Market Screens Board is very good as well, though more narrow in scope.


Cheers K
Print the post Back To Top
No. of Recommendations: 0
Actually, I had not been to that board yet. Thanks for the tip.  I checked it out, but had no idea what anyone was talking about. I guess I'll have to do a lot of reading to get up to speed.

Thanks for the tip.

Jon
Print the post Back To Top
No. of Recommendations: 0
Jon,

What database are you using for your backtesting? Unless you have a database that includes all the symbol changes, delistings, etc., the period between 2000 and 2002 might look even worse than you think.

I'm very interested to see that your system makes good money on the short side. I have tried hard to come up with a good short system, but have never been able to develop anything I'd be willing to trade.

ges
Print the post Back To Top