Indicator Bullshit - page 2
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I have it in my shelf, it's a decent book athough I think (and the author himself basically states it himself) that his tests are too simple. In fact he states that some other studies have found correlation with neural networks and some more complex combinations of indicators - source list on that book is a goldmine of good research.
Ok, I'll re-phrase it to be exact, indicators used in isolation are statically insignificant as buy/sell signals.
Also I think the stock market is not the best market to test technical rules on, stocks are just too random when compared to forex or futures.
Are you sure about that? All the successful automated traders I know of are stock traders simply because the markets are less efficient. The forex markets are very efficient, hence the difficulty in finding an edge. This is also the reason the cost of entry is so low, more meat for the grinder.
Are you sure about that? All the successful automated traders I know of are stock traders simply because the markets are less efficient. The forex markets are very efficient, hence the difficulty in finding an edge. This is also the reason the cost of entry is so low, more meat for the grinder.
Of course I'm not sure
Although I have traded stocks in the past but it was not daytrading nor technical trading, more based on fundamentals.
For technical trader fluctuations in stocks tend to be too random, for example there is no technical way of knowing if a management has been hiding losses or some other sudden unexpected event occurs.
Indexes are other thing though.. Much less fluctuations.
Forex is a good market because countries economies change far slower than individual companies business situations do, therefore the trends tend to be much much longer.
By the way I found it a bit funny that the guy who wrote the book we were discussing detrended the data before testing - when he did that he basically admitted that there is something called trend that will affect your results.
Bullshit???
Indicators are nothing more than a visual representation of an equation utilizing specific data derived from price.
I'm not screaming... I just wanted people to notice that statement. Notice I did not mention things like, clairvoyance or omniscients. Saying they're bullshit is like saying all calculators have no use or significance. This issue gets thrown around so often it's nauseating. Stop looking for the crystal ball, and start utilizing the "Bullshit" tools that are available. If you don't want to use them... then don't. Nothing wrong in either case.
Peace out,
MM
Common sense....not as common as you think
Mr Marketz - Indicators are nothing more than a visual representation of an equation utilizing specific data derived from price.
Hoorah, finally a valid comment from somebody whom knows what they are talking about! Mr Marketz could not have stated it more succinctly.
I laugh every time somebody says they are going to create a "price action derived EA that does not calculate it's entry via indicators". If they thought about it for 10 seconds (or less) they would see that ANY math once applied to price in order to attempt assume it's future direction is merely creating a new indicator. Of course, this math TRULY becomes an indicator by definition once you build it into a visual shell that can be attached to the chart.
Yes, there are NO magic indicators that will do the job of trading for you but as tools in the hands of a skilled tradesman, they are invaluable. ANYTHING that even gives you just 1% of an edge (and this industry is about probability), then it is 1% worth of gold not to be scoffed at.
That particular article did get one thing correct - The importance of money management. You can have the most rubbish of systems and yet if you know how to handle lot sizing, RRR and control your environment when a string of losses ensues (it will happen), you will make a very profitable living by trading. 98% of people on forums are seeking that 100% accurate system that will remove the stress from this job only to end up wasting many years doing so and still burn their money because they simply don't what trading is all about. It doesnt matter to me, quite a few will disagree with this post and I just don't give a damn.
Indicators DO work, markets are inefficient (most people that try to prove they are efficient are poor) and as traders you must learn to take losses!
Happy trading.
FxNorth
If you really want to know about the statistical significance of indicators read 'Evidence-based technical analysis', you can get it from amazon. The guy in the article is just trying to push his stuff but you needn't dig very far to find actual published evidence that indicators are indeed 'bullshit'.
EBTA is a good book, and one I often reccommend, but I dont think the author really "gets it"
He falls into the trap of believing indicators are some way predictive, with the inevitable confusion that causes. He'll never be able to prove an indicator is predictive, because they arent !, I'm not sure why even even tries too.
You also have to admit that the author of the book does take rather a simplistic approach to exits, I dont of know many traders who close after a set number of bars irrelevant of profit or loss, its not a particularly sophisticated or practical exit strategy.
I actually think that almost all indicators are useless unless used in special situations like predicting if the market is peaking when you want to make your stops larger and that's in fact predictive usage of indicators
Lets consider a really simple example. Lets say that you believe trends exist, and price goes up and down. You stick a simple indicator such as ATR on the timeseries, and you fiddle about with the indicators parameters.
Maybe you read somewhere that settings of X are optimum for this indicator, maybe you look at the charts, and you fool yourself that settings of Y are optimum, it doesnt really matter, its the equivelent of measuring the distance between two points as kilometers or inches, very different answers, but both fundementally the same thing.
You've decided that these trends really do exist. You notice that price kind of wriggles around a bit. You can probably even work out the basic market mechanics as to why this happens, identifying the different participants, where they're likely to get in and out etc.
You do a bit of basic research and you find out that when the markets trending for your particular instrument and timeframe, price is always constrained within a channel of X ATR, 90% of the time price is constrained within Y ATR, 50% of the time within Z ATR etc.
Now put some other pieces of the jigsaw in place, the ability to identify a trend, a basic understanding of how news drives the market etc. Lets assume you take a position in the market, and you use your basic knowledge of ATR to assist in stop placement (along with other factors)
Is that really using an indicator in a predictive way ?
I'd argue that the indicator isnt predicting anything, its just providing a confirmation that the market isnt behaving as I would like it too. Therefore non of the assumptions upon which my methodology is predicated are in play.
He falls into the trap of believing indicators are some way predictive, with the inevitable confusion that causes.
IMHO the principles of the study are done very well and he's on the spot on his conclusions, the problem with this particular research is that in todays markets it no longer applies - if you apply basic TA like breakout rules, moving average crosses and so on to the 1920s-1970s daily data and they really work.
The book is not just about the indicators, it's about TA rules for applying them and if those are not meant to be predictive (ie. have some correlation to how market reacts in the future, long or short term) then what would be the motivation to use any indicators (including price) at all?
I already said this but IMHO the biggest thing in the book is that he should have combined trend indicators to rules and not detrend the data. Detrending kills the usability of that particular research in real world situations.
edit: and I actually think that almost all indicators are useless unless used in special situations like predicting if the market is peaking when you want to make your stops larger and that's in fact predictive usage of indicators
Indicators are nothing more than a visual representation of an equation utilizing specific data derived from price.
I'm not screaming... I just wanted people to notice that statement. Notice I did not mention things like, clairvoyance or omniscients. Saying they're bullshit is like saying all calculators have no use or significance. This issue gets thrown around so often it's nauseating. Stop looking for the crystal ball, and start utilizing the "Bullshit" tools that are available. If you don't want to use them... then don't. Nothing wrong in either case.
Peace out,
MMI agree. Although take it easy on our eyes dude.
I agree. Although take it easy on our eyes dude.
- for some people it helps with better understanding, so they have to use big letters...
________
inds - just graphical representation of the numbers, data (statistics, etc) according to which it's not lying - our interpretations, conclusions and expectations can be wrong but - let's blame average - 4 it's not true average...
... they do what exactly we told them to do - and they always wrong...
EBTA is a good book, and one I often reccommend, but I dont think the author really "gets it"
He falls into the trap of believing indicators are some way predictive, with the inevitable confusion that causes. He'll never be able to prove an indicator is predictive, because they arent !, I'm not sure why even even tries too.
You also have to admit that the author of the book does take rather a simplistic approach to exits, I dont of know many traders who close after a set number of bars irrelevant of profit or loss, its not a particularly sophisticated or practical exit strategy.I'm not saying EBTA is perfect, but it is somewhat more well thought out than most information you see on indicators. I actually thought the information on data mining bias was more valuable and interesting, very applicable to what goes on around here.