Topic: Random Walk - or - Is Technical Analysis Hocus Pocus?

Hi,

I have a question that is not so much about this software but about the results you have found with all this back testing. According to random walk theory, there are no predictable, repeatable patterns and the price data is just random therefore making technical analysis a waste of time.

My question is, after all your testing have you found this to be the case? Or is there patterns in the price movements and there is a possibility of creating a profitable mechanical tech/a system?

I suppose it is more of a philosophical/academic question but would be greatly interested in your theories.

Kind Regards,

onemind

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Just thought i would add a post of mine from another forum:

--------------------

Anyway, according to random walk theory and the efficient market hypothesis day to day price movements are random. Therefore making all technical indicators useless as past prices have zero effect over the probablilities of future price movements. Then tech/a people counter that by saying the markets cant be random because they trend and yes, the trend is your friend. Then that gets countered by a demonstration of a random data model with stochastic distrubution, fat tails and other statistical jargon showing clear trends in the random data.

Then in comes mandelbroit with his fractal theories ect but they are in such beginning stages to be un usable for day to day practical purposes.

Now, i figured that the best way to settle this would be to back test each indicator over years and years worth of historical data across many different markets to see if the indicators do have any predictability elements. I then found this book:

http://www.amazon.com/gp/product/customer-reviews/0470008741/sr=1-10/qid=1177056540/ref=cm_cr_dp_2_1/104-7048091-1143941?ie=UTF8&customer-reviews.sort%5Fby=-SubmissionDate&n=283155 &qid=1177056540&sr=1-10

Which did all the statistical work for me and it is shown that 6400 technical indicators turned out to be useless.

So how do people make money with tech/a? The answer is that millions of people use tech/a with proper money management. Out of that many people doing it there are bound to be a small percentage of winners from luck and a large amount of losers which would make sense of the often quoted statistic that 90% of traders lose money.

So is it fair to conclude that successful chartists are just luckier than their fellow losing chartists?

---------------

My question to you guys is, do you trade yoursleves? Or did you figure the software business is more profitable? smile

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

So i guess this argument just makes this software useless? Other than for proving that t/a = astrology smile

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

The sole purpose of FSB is to give you an easier way to find your truth!

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Well, i doubt the truth is subjective. T/A indicators either work or they dont.

Do we really need 1 million monkeys testing the same indicators over the same data and hope one of them comes up with a different result? Isnt that like the definition of insanity?

smile

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Dear friend,
I have an idea about a trading system.
It is a very simple one. You need to measure only two parameters of the market - the direction (speed) and its inertness.

The rules are easy.
   Buy/sell when the price rises/falls with speed and inertia greater than certain minimal levels.
   Close when either the speed or the inertia drops.

That is all. You have the main idea now. So, try this technical system and find the answer of your question!

Opss!
Please, be a real friend and tell us does it work or not? smile wink

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Sounds good Popov.

The trouble is, the speed at any moment is random and can change at any time.

Do you trade yourself?

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Yes, I trade.
And to answer to your further questions.
Yes, I use technical analysis.
I neither lose too much to reject the T/A nor earn enough to confirm it.

I'm not the one, ho will tell you to use the T/A or not. You are free to decide alone!

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Thanks agian Popov.

I think you are by far the most helpful trader around smile

You really should go commerical with this software when its finished. Its a great idea for the upcoming boom of retail forex traders with he new tech coming.

Cheers

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

You might become convinced as to the efficacy of using a system and tips from Phenomenal Successful World Class Expert Traders here who have proven their abilities in competition with each other. >>>

http://www.worldcupadvisor.com/

Cogito Ergo Sum

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

No market in the world is random. They are influenced by emotions. Market always reacts more than actually correct amounts. So both TA and FA work in all markets. YOu can make money if you use TA appropriately. Too many indicators will make your life miserable.

No indicator is perfect. This does not mean that the market is random. The market usually moves 70 to 200 pips a day. Out of this 40 to 60 points are usually random noise. The rest is not random. waqtch out for that. Read 4 x 1 strategy to get a better idea of markets. You can test it out in forexsb. You will get a win loss ratio pf 90% and around 2000 pips in a few months. You can consistantly aim for 10% returns a month.

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Ola!
What is that magic strategy viswadatta!
Can you post it?

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

I have already posted it under logic and strategies heading

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

I've posted some too, and I even check to see if a walk-forward region is still havig the equity crawl upwards with similar small DD, think I've posted 2 or 3 I found so far.  I'm in habit of rejecting FSB results that don't walk forward for a reasonable period of time compared to amount of data in our FSB, it's simple really, market > horizon > chop-off data, generate, un-chop to see overall equity curve (i.e., equity not falling down at end or with big DD?), save if good both in training area and validated walk-forward.  Think I have 4 saved so far.  However, we do need Popov to have a validation region feedback option during generation with selection of DD we desire.  Even before that we need simple to do itemsthat are check-box suppress of a few bummer strategies from coming "to the top", like bad DDs, multiple lots, and orders during bars types, this will avoid all the all-too-often recurring encounters of the strange strategies that are mostly un-doable.  Please write Popov of these needs, very important.

jerry

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Many traders are convinced that even without using indicators, and using random open/close, trades can be profitable using the correct Money Management strategy.

Read this  >>>  http://www.tradingmarkets.com/.site/Swingtrading/commentary/satinterview/10142005-46432.cfm

And this  >>>   http://www.moneybags.com.au/downloads/extracts/0471297224.pdf

And  here  >>>  http://64.233.169.104/search?q=cache:wiOjmcJQzk8J:forexmadeeasystatistics.blogspot.com/+random+profits+%22money+management%22&hl=en&ct=clnk&cd=6&gl=us&lr=lang_en


I hope this helps convince everyone that Money Management is THE Most Important Requirement to stay in the game. If you are one of the many who experience an unlucky series of early losses, as happens to most, then you lost most of your money and are now out of the game, unless you are lucky enough to catch a winning streak to recover your losses. But then, you are back to where you started and facing the same odds of losing it all again.

Good Intelligent Trading outcomes to you all!

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

onemind wrote:

Just thought i would add a post of mine from another forum:

--------------------

Anyway, according to random walk theory and the efficient market hypothesis day to day price movements are random. Therefore making all technical indicators useless as past prices have zero effect over the probablilities of future price movements. Then tech/a people counter that by saying the markets cant be random because they trend and yes, the trend is your friend. Then that gets countered by a demonstration of a random data model with stochastic distrubution, fat tails and other statistical jargon showing clear trends in the random data.

Then in comes mandelbroit with his fractal theories ect but they are in such beginning stages to be un usable for day to day practical purposes.

Now, i figured that the best way to settle this would be to back test each indicator over years and years worth of historical data across many different markets to see if the indicators do have any predictability elements. I then found this book:

http://www.amazon.com/gp/product/customer-reviews/0470008741/sr=1-10/qid=1177056540/ref=cm_cr_dp_2_1/104-7048091-1143941?ie=UTF8&customer-reviews.sort%5Fby=-SubmissionDate&n=283155 &qid=1177056540&sr=1-10

Which did all the statistical work for me and it is shown that 6400 technical indicators turned out to be useless.

So how do people make money with tech/a? The answer is that millions of people use tech/a with proper money management. Out of that many people doing it there are bound to be a small percentage of winners from luck and a large amount of losers which would make sense of the often quoted statistic that 90% of traders lose money.

So is it fair to conclude that successful chartists are just luckier than their fellow losing chartists?

---------------

My question to you guys is, do you trade yoursleves? Or did you figure the software business is more profitable? smile

Cogito Ergo Sum

Re: Random Walk - or - Is Technical Analysis Hocus Pocus?

Here is an interesting backtest using Random Entry + Bollinger Bands Exit System  >>>

http://www.compuvision.com.au/Examples/HolyGrailPart1.htm

Cogito Ergo Sum