Is any prediction doomed? - page 17

 
FAGOTT:

cardiogram?
Coprogram.
 
DDFedor:
Do not use the word "forecast". Use the word "conclusion". "Prediction" is uncertainty, "the will of chance". "Inference" can be formalised.
A prediction is the result of analysing a particular set of data.
 
FAGOTT:


If you have found a pattern in the financial market that has worked 100 out of 100 times before, does that mean that in the future this pattern will work with the same probability?

Even more narrowly - do you know a pattern in the market that works 100% of the time? A pattern that does not have losing trades over a long period of time?

Tell me. Why, you are focusing on the fact that one "dependence" you find should immediately give you a constant profit? These tiny "addictions" are always intermingling with each other, you always have to look for a combination. Yes, there is a condition that has been repeated three times a day for many months now. But this is only a very small clue that will certainly be included in the working system on the way out. There are many such "leads". But it is not these "leads" that need to be looked for. You should be looking for the origins. And the sources are the SAR lines. The ways of their construction and passage are the basis for the calculation of the price movement.
 
DDFedor:
Tell me. Why, do you emphasize that one "dependence" you find should immediately give you a permanent profit? These tiny "dependencies" are always intermingling with each other, you always have to look for a combination. Yes, there is a condition that has been repeated three times a day for many months now. But this is only a very small clue that will certainly be included in the working system on the way out. There are many such "leads". But it is not these "leads" that need to be looked for. You should be looking for the origins. And the sources are the SAR lines. The way they are drawn and their passing through is the basis for calculation of price movement.


))) It is not about that at all.

All this time it was about this phrase and others like it - "Generally speaking, predicting in the financial markets is useless. Only the search for patterns.....".

The whole argument was about the fact that no prediction is possible without the identified patterns and the identified patterns are useless without prediction for trading.

PS lines - support/resistance? I am extremely skeptical of TA, and certainly of its graphical methods....

 
FAGOTT:


If you have been observing the pattern for a long enough period, what prevents you from using it? The feeling that the money earned in this way is "wrong"?

Victor Nidderhoffer is a prominent trader and multi-millionaire. He describes in his book the correlation between a country's level of wealth and the length of its butts. He uses this.

Also in his book he describes the correlation between price movements in the financial markets and the migration of elephants in Africa. And he uses this too in trading. And successfully


Screw the gurus :)

If someone likes it, he uses it as long as he understands what he says.

As for the search for patterns, there are only two ways - statistics and logic (in this case, trading). And they complement each other.

 
now usually in the sense of a probabilistic judgment about the future based on special scientific research (from this source)
 
It is not certain that it is doomed, but whether we know how to make predictions and whether we have laid the right assumptions in the basis of scientific research. It is necessary to search for regularities that increase the probability of the forecasts coming true. The search for regularities should not be opposed to forecasting or try to replace one concept with another.
 
FAGOTT: I am extremely sceptical of TA, and certainly of its graphical methods....
What are you not sceptical about?
 
FAGOTT:


Fuck, I'm getting these explanations on my fingers...

OK, I'll try it - I'm crossing an intersection at a green traffic light. Why? Because when I cross on green, I estimate my chances of getting hit by a car are lower than when I cross on red (probability, bollocks). I have discovered this pattern and I predict that my crossing the intersection at a green light is more likely to end successfully than my crossing at a red light.

Most pedestrian collisions statistically occur at pedestrian crossings.