Technical analysis - A scam for beginners? - page 10
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
Although I don't have as much experience as Sergei (Farnsworth), and I've been doing technical analysis for only three years (since I learned about FOREX). Nevertheless, I've made some observations (mainly with strategy tester) and believe that TA is not as hopeless as many people think. Here are my main claims:
1. technical analysis has a very limited predictive horizon. It is only realistic to predict the movement a few bars ahead and no more.
2) It does not make sense to make calculation data windows of more than three five figures. Roughly speaking, what was 5-10 or more bars ago is irrelevant to what is happening now.
3. It is important to understand that the so-called TA figures are clearly traceable on anything from an accumulative coin flip chart to temperature dynamics. For example, if a clear five-wave Elliott wave pattern showed up only on markets and never on SB or weather charts, I would be the first to throw stones at those who would claim that Elliott waves don't work. But unfortunately Elliot waves are meaningless as they appear on anything and since they have no meaning, they cannot work. This also applies to all other TA formations like double top, triangle, head-shoulders, etc.
4. Before using a technical indicator, you have to clearly answer yourself the questions: "What exactly this indicator shows?", "What exactly I need to see in the price movement?", "Why exactly this indicator shows price deviations which I need to see?", "In which cases this indicator will give clear signals and in which cases it will be wrong?".
5. Filters. I do not think it is necessary to use more than one filter in any trading system. In special cases you can use two of them. As a rule, several indicators are selected based on the principle: "It showed a nice reversal here" and used together. As a result the non-working parts are gathered in a non-working system, which is approximated very well (which means the fitting on the history). Past that, any slight deviation breaks the whole TS.
6. Simplicity of approach. My belief that the simple moving average is not much worse than the integral of the linear regression slope or some polynomial smoothed price series. The effects of deviation of price movement from the normal SB movement should be visible with any, more or less adequate analysis tools, even visual ones. The effects should be observed with the help of any tools and any approaches.
From the above it becomes clear that TA has its roots in the nonsensical science of observing non-existent patterns. And to this day, more than anywhere else, these beliefs are strong in it. TA is like modern chemistry. It came from alchemy, a science whose aim was to find a way to turn stones into gold. And although alchemy is considered to be a pseudoscience, chemistry has made tremendous strides in mastering the world. It has made many interesting discoveries and changed the world around us dramatically. So has TA. Today TA is a precise and efficient science, a world ruled by automatic trading systems, exact calculations and simple mathematics. But along with it there are still strong old atavisms, the same "alchemy" of traders of the past. Books are still being written on it, and it is mainly from this old TA that our twisted perceptions of the present and the modern are formed.
At least don't call TA a science. There is a scientific discipline, the sociology of knowledge, which gives a clear and specific definition of science. And the criteria for defining science.
So read Taleb. TA according to these definitions is a typical example of pseudoscience/false science. This has already been written about on this forum. As, by the way, so is alchemy.
1. Why a few bars in particular? How many bars in particular? Which bars? A few daily bars? Where did you get that from? Predict.
3. what other five wave process? That is, wherever there is a wave process and the number of waves is accidentally observed on a certain segment equal to 5, it is a "five-wave"? So I will find two-wave and three-wave and so on.
At least don't call TA a science. There is a scientific discipline, the sociology of knowledge, which gives a clear and specific definition of science. And the criteria for defining science.
So read Taleb. TA according to these definitions is a typical example of pseudoscience/false science. This has already been written about on this forum. As, by the way, so is alchemy.
6. Simplicity of approach. My belief is that a simple moving average is as good as a linear regression slope integral or some polynomial smoothed price series. The effects of deviation of price movement from the normal SB movement should be visible with any, more or less adequate analysis tools, even visual ones. The effects should be observable with any tools and any approach.
Have you even tested them? To say they are meaningless.
What does this have to do with testing? They, these figures, really stick out on all fluctuating charts, even random ones.
On a market chart, they only have more power due to the psychology of the crowd, i.e. you and me.
what does this have to do with testing? They, these figures, really stick out on all fluctuating charts, even random ones.
On a market chart, they only have more power due to the psychology of the crowd, i.e. you and me.
Right, we haven't tested them.
Well, they work on temperature charts too. Anything with a trend component works. And no psychology. Fucking inertia.
What does this have to do with testing? They, these figures, really stick out on all fluctuating charts, even random ones.
On the market chart, they only have great power thanks to the psychology of the crowd, i.e. you and me.
You can find some repeating shapes on almost any chart of a random process. The main thing is to have enough sampling.
The problem for trading is that they are identifiable only after they are finally drawn and therefore useless for forecasting
We're having a conversation here. The boss won't be happy, there was a clear instruction to lay out the stats.