Optimizing or curvefitting

>>I read Pardo's book in 1994 or 1995. It was one of the most useless books on trading that I have ever read. Of course, this is a personal opinion.<<

Your comment reminds me of a Day Trader that I had chatted with. He mentioned that back testing was useless and never worked.

I have to disagree. I have been trading for almost 2 years... A newbie. My first year of back testing, I only found systems, that lost money, or after trying what I thought were good, ended up producing poor results.

The point I would like to make, is after doing lot's of back testing for 2 years, I am able to back test a system, and then trade that system in Real Time. The thing that I like, was I am able to duplicate my back test results in real life trading. This is much tougher than it looks.

The trick is learn as much as you can about developing trading systems, analysis, etc.

The point, is just because something does not work for you, does not mean it is useless. However, I would not use something that did not work for me.

I have re-read many trading books and have learned a lot the 2nd time around.

This is the case for me about Pardo's book.

Just my 2 cents...

larryTAKEOUT@seldin.net
 
Quote from hypostomus:

...I am just a happy idiot. By no means did I mean to disparage the joys of tediously speculating on market action bar-by-bar. That is exactly what I do to distract myself from impetuous action while I am in a trade. I test each delusional idea as it occurs to my febrile mind, which wants only to escape from the tension of holding a position. I have tested scores of worthless ideas in the past two years and have only five (apparently) working systems to show for it. I never cease to be amazed at how certain I am that this or that pattern works, only to find that it is a loser overall, and that memory retains only the rare wins.
More...

Bonjour hypostomus,

Happy to read some true wisdom. Don't often find this at ET's. It seems to me that you don't go much with either "curvefitting" or "optimizing". That's only for fellows who have not yet attained your wisdom.

Nononsense ain't kidding this time!

Be good,

nononsense
 
Quote from indahook:
Obviously one of the most important things to consider before
running a trading system is determining the type of mkt you are
trading i.e trending up, down or sideways. Which leads me to this question.
:D
More...

I am presently look at a common technique. Only trading in the direction of the long term trend.

I have good system that I have been trading. If we assume, that I am able to define the longer term trend, then I was able to test, trades that were taken only in the direction of the long term trend.

My testing for my own system, is that I did not produce as much profits doing this.

It seems that even trying to eliminate the whipsaws, also filtered out good signals.

So the point, is that yes, there are systems that probably do much better trading in the direction of the longer term trend. I would just test any system you are trading, to "prove that out" for each individual case.

larryTAKEOUT@seldin.net
 
Quote from seldin:

I am presently look at a common technique. Only trading in the direction of the long term trend.

I have good system that I have been trading. If we assume, that I am able to define the longer term trend, then I was able to test, trades that were taken only in the direction of the long term trend.

My testing for my own system, is that I did not produce as much profits doing this.

It seems that even trying to eliminate the whipsaws, also filtered out good signals.

So the point, is that yes, there are systems that probably do much better trading in the direction of the longer term trend. I would just test any system you are trading, to "prove that out" for each individual case.

larryTAKEOUT@seldin.net
More...

Brilliant idea Seldin,

I found out this only works well if you know how to find the "long term trend".

Hope this helps. :D


nononsense
 
Wisdom is a relative thing, and I am relatively safe from it, except when one of my posts is juxtaposed with one of Harry's! If I were a huckster I would definitely hire him to make the sales pitches? Whaddhesay?

Regarding optimization, I more than ever adhere to my early rule of no more than seven numerical parameters in a system. There is no way that such a system can be curvefitted. However, I do optimize the hell out of a system using those seven, so the result can still be a fit to random data. Only time tells.
 
Quote from nononsense:

Brilliant idea Seldin,

I found out this only works well if you know how to find the "long term trend".

Hope this helps. :D


nononsense
More...

Agreed, thats why my directional filter is very vague.

My definition of a Bull, Bear or sideways trend must be the same as everyone elses or the system wont work. I want to be the sheep that hides in the middle of the pack and barely has to walk from all the pushing behind me. Then when I sense the packs frustration in carrying my weight slip out the side un detected.
 
Quote from indahook:

Agreed, thats why my directional filter is very vague.

My definition of a Bull, Bear or sideways trend must be the same as everyone elses or the system wont work. I want to be the sheep that hides in the middle of the pack and barely has to walk from all the pushing behind me. Then when I sense the packs frustration in carrying my weight slip out the side un detected.
More...

Indahook,

Your picture of a "sheep that hides in the middle of the pack" is very fitting. The difficulty for the sheep is to KNOW where it is going so that it can decide on when to place its longs and when its shorts.

nononsense
 
Quote from hypostomus:

..I have tested scores of worthless ideas in the past two years and have only five (apparently) working systems to show for it.
More...

Did you trade 11 systems before? Are you implying the other 6 not working anymore? That could be very valuable experience to share with us.

:confused:
 
...your memory is entirely too good. Back when I was attempting to find a trade in NQ every morning I had developed twelve systems to exploit what seemed at the time to be patterns or structures in the first hour. They kicked in sequentially starting with the open, and I used later calls to reinforce or reverse earlier calls.

Of the twelve, eleven survived. The "structures" turned out to be evanescent, and the volatility-based systems succumbed to volatility reductions and a general truncation of the morning range. The reversal calls turned out to be unreliable.

I subsequently developed two additional morning systems which trade much less frequently and which tend to be mutually consistent. I also added an afternoon system and an all day system.

Based on advice from you and others, I also chose optimization parameters which produced slightly suboptimal results to get more robustness. Those optimization parameters tend to be more stationary over time.

Best regards. - Mike
 
You must log in or register to reply here.
Back
Top