| Author |
Message |
     chart_rider
Member
Username: chart_rider
Post Number: 152
Registered: 01-2005Rating: 
Votes: 3
|
| | Sunday, January 29, 2006 - 05:22 pm: |  
|
Good evening all,
For a considerable length of time I have been developing a trend following system for the purpose of position trading (from the long side).
There's no rocket science in the system, just some simple rules that have passed rigorous backtesting across different markets (shares + futures).
One particular section of "The New Market Wizards" talks about using hypnotic therapy to program the subconscious mind into accepting and believing what the conscious mind already knows. Thus, given the conscious mind is happy with a trading method based upon good statistical outcome, the subconscious mind may not have accepted this and gets in the way of emotion free trading in strict accordance with the plan.
How the mere mortal interferes with the trading method!
"Mind over matter" proves itself to be a worthless and undesirable concept in trading. Apparently trading is not the same as bending spoons!
I trade for a while, perhaps several weeks or so, then analyse the returns, which appear to be a bit low. As the trading method is based upon a series of rules, performance can be measured by the level of adherence to the rules.
So, the same period is run through the backtest, the computer diligently follows the rules, just as it was programmed to do, and, for the same period, the computer exceeds the return achieved by its mortal master by several hundred percent!
This is not an isolated case.
What is happening here?...........This is what happens:
1. Mortals read magazines about the market, and become biased by what is read - the computer does not suffer this problem as it can not read.
2. Mortals can see charts that almost, oh so almost, fire off an entry trigger - the computer does not suffer this problem as it can not see.
3. Mortals think, a stock is pumped up by economic forecasts in magazines (of course resource stocks are the best buy - are'nt they?) AND the stock, oh by a whisker , just missed out on an entry trigger, so why not give this stock the benefit of the doubt and buy it - the computer does not suffer this problem as it can not think.
The problem is clear, performance is hindered by the following human capabilities:
The ability to read,
the ability to see, and,
the ability to think.
If a trading plan is in a state of high development, one must accept that the thinking is done! Must one put the thinking to rest when trading and just follow the blarsted rules?!
But, if the computer can do so well at trading in accordance with the plan, then why do I need to employ my eyes and brain for trading - perhaps it's better to let the computer trade!
My thoughts (I'm allowed to think at the moment as writing this post is not actually trading) is to not look at the charts to decide on possible entries - let the computer tell me what to enter, that way I can't mess with the purity of the method.
I have added a new rule to my trading method - "the computer shall be my subconscious".
In the mean time, the blind fold comes off only at the end of the day when it's safe to analyse and ponder.
CR
|
spider
Member
Username: spider
Post Number: 2379
Registered: 10-2002Rating: N/A
Votes: 0
|
| | Sunday, January 29, 2006 - 05:41 pm: |
|
CR,
excellent article.
I'm not sure that I entirely agree, but it is difficult to fault the logic.
The less I fiddle with my trading, the better I get.
Keep us posted, I would love to hear more.
spider.
Are machines better at this than we are? I parked the car, sat quietly and just though about where this might lead.
|
davkell
Member
Username: davkell
Post Number: 346
Registered: 07-2004Rating: N/A
Votes: 0
|
| | Sunday, January 29, 2006 - 07:20 pm: |
|
Makes a lot of sense CR. When I first got into this wonderful world we call 'trading', it was with a 'Grey' Box system. For $8000 (ish), I didn't get one, not ONE chart! The system was based on something like 200 odd formulae (probably just our standard technical indicators) that then gave the user an 'overall' valuation of the stock via several screens in percentage type values.
Without going into the complexities of the indicators, the whole program allowed you to understand what the underlying market for a particular stock was doing, just based on numbers! Hence, there was no interpretation of the chart, no second guessing whether the pattern had occurred, was occurring or almost there!!!! 50% is 50%. 49.9% is NOT 50%! Hence, your rules were followed to the letter, or in this case the number!
CR, if your computer spits out such a bigger difference in trading results, then I think the mechanical, do as the computer says, approach is the way to go.
If I had the time and the Macro knowledge, I would like to switch my system to purely numbers. Put in the stock prices, the calculations for my indicators, and the formulae to recognise my trading signals. Hence, trade the numbers, not the picture (no matter how you see it!)
"Trade Your Way To Financial Freedom" - Van K Tharp
"Manage the downside; the upside will take care of itself" - Donald Trump
|
msparks
Member
Username: msparks
Post Number: 308
Registered: 10-2004Rating: N/A
Votes: 0
|
| | Sunday, January 29, 2006 - 09:45 pm: |
|
Guys
How do you know your order will be filled when backtesting.
Does the backtesting enter on close or open price ?
Do you find most of the up move occurs between the close and open yet a lot of people advise to waite half an hour until the open "action " is over ?
If you automated a system, how would the system see the depth to guarantee a fill at the open ? if this was the plan.
Cheers
|
chart_rider
Member
Username: chart_rider
Post Number: 153
Registered: 01-2005Rating: N/A
Votes: 0
|
| | Monday, January 30, 2006 - 06:42 am: |
|
msparks
Personally I also prefer to stay clear of opening action, I just can't make sense of it and have lost too much in attempting to exploit this volatile period.
If trading liquid stocks, eg ASX 200 or 300, picking stocks that trade above a certain level of daily turnover, the market depth becomes less of an issue.
Configuring turnover into the backtest eliminates stocks that could have depth problems. In my case trading is based upon closing, but opening prices could be used in a backtest with participation in the opening auctions during real trading resulting in close correlations with the results.
Davkell
The question is whether it is possible to exercise the discipline to follow a system without knowing what's inside it.
With a self developed system, one knows exactly how it was tested and how it needs to be applied. Tests can be carried out with varying parameters to ensure robustness, as opposed to a system that was curve fitted with tight parameters - one just doesn't know with a bought system.
CR
|
chart_rider
Member
Username: chart_rider
Post Number: 157
Registered: 01-2005Rating: N/A
Votes: 0
|
| | Tuesday, February 28, 2006 - 07:22 pm: |
|
Spider, and Forum
Well, it's been a month now since handing my sub-conscience over to the computer for safe keeping. The experience has been interesting and encouraging.
Some of the positives of this new found method include -
Expeditious analysis. The computer takes 14.9 seconds to scan the ASX300 and provide a watchlist for the next day. From this point a stock only needs to do one thing to be scooped up.
This leaves more time for developing strategies and less time spent developing opinions. After all, developing strategies is half the fun of trading.
Elimination of the squirm effect. No need to agonise over a chart, seeing different things upon each glance. The computer's word is final.
The comfort of company. If a particular trade goes against me, it has gone against the computer also, so I'm not alone.
A well defined bench mark to strive for. The computer is really very good at this stuff, and to match her grade takes complete focus and discipline.
Is the holly grail really just learning to mechanically follow a system?
CR
|
stevo
Member
Username: stevo
Post Number: 338
Registered: 01-2003Rating: N/A
Votes: 0
|
| | Tuesday, February 28, 2006 - 09:12 pm: |
|
Chart Rider
Is the holly grail really just learning to mechanically follow a system?
"We have not even to risk the adventure alone, for the heroes of all time have gone before us. The labyrinth is thoroughly known. We have only to follow the thread of the hero path. And where we had thought to find an abomination, we shall find a god; and where we had thought to slay another we shall slay ourselves; where we had thought to travel outward, we shall come to the center of our own existence; and where we had thought to be alone, we shall be with all the world."
But not just "a system", you better use one that you know works. I like the idea of the computer as your companion in the search.
stevo
Quote from - Joseph Campbell's The Power of Myth Vol. 1: The Hero's Adventure.
Also quoted in Van Tharp's book with the ridiculously long title, chapter 1.
|
chart_rider
Member
Username: chart_rider
Post Number: 158
Registered: 01-2005Rating: N/A
Votes: 0
|
| | Thursday, March 02, 2006 - 11:39 am: |
|
stevo
A worn path is truly known only by the heroes themselves, and one who has not followed the thread is not yet a hero.
Uncontaminated senses experiencing unique inspiration from a previously trod path will lead to personal adventure from which one may become a hero within.
...........CR, "Every one is a hero", 2/3/06.
The market way has been worn smooth by those before, and yet is not smooth to the new adventurer. Only the properly prepared hold the option of prosperity, as you said Stevo, you better use a system that you know works - I couldn't agree more.
CR
|
|
