Probability. - page 4

 
Uladzimir Izerski:

My opinion. Analysis of the current moment only.

I believe it is not necessary to estimate the probabilities of the current moment. It is better to estimate if the current moment falls under the signs of a good deal. And the results of many trades will show with what probability current moment estimation is correct. And it is more productive to look for criteria than probabilities.
 
Yuriy Asaulenko:

Total nonsense. The probability of winning in poker is 1/9-1/6 - that's only 15% or less. Nevertheless, good players are in profit, although their winnings fall within the specified probability range. I am a good poker player myself and my winnings don't exceed 20% of games played (I can show you statistics from my web-site); nevertheless my profit is rather good.

By the way, poker techniques are very well applied to the market.


This coming from a guy who wrote an hour ago that you can make money on random walks, although all theoretical books state the opposite.

I don't know how to play poker, but I know you're not making any sense.)
There's a 50% chance of winning the game if two players are playing.

how hard is it to explain maths to people...


 
danminin:

And this is coming from a man who wrote an hour ago that you can make money on random wandering, even though all the theorist books say otherwise.

i don't know how to play poker, but i know you're talking rubbish.)
There's a 50% chance of winning the game if two players are playing.

how hard is it to explain maths to people...

This is from the cycle - I haven't read it, but I know it.)))

I never wrote or said that you can make money on random wandering. Read carefully.)) And learn some maths, by the way, before explaining anything to anyone.))

 
Uladzimir Izerski:

how do you calculate the probability?

Let's say a bad nonfarm is released and you know from history that the chart goes down 70% of the time after such news.
you say "there's a 70% probability the chart should go down now!"
that's it! you use historical data!

or do you want to use technical analysis to calculate probability? give me an example of such a calculation.

 
Yuriy Asaulenko:
I believe it is not necessary to estimate the probabilities of the current moment. It is better to estimate if the current moment falls under the sign of a good deal. And the results of many trades will show with what probability the estimate of the current moment is correct. And it is more productive to look for criteria than probabilities.

One does not preclude the other. But one must always assess the probabilities of the current moment.

News, important data, etc. have to be taken into account however you want, and they may not match the historical results.

 
Uladzimir Izerski:


And in fact, new conditions are being created in the world and consequently in the market that will create a new version that does not correspond to the historical image.


My opinion. Analysis of the current moment only.

What conditions? How can the current moment be counted?

Unless you set the coefficients.

For example, there are 3 news items during the day.
Bad news, set the odds at 0.9.
A good interest rate change. Put odds at 1.1.
A bad inflation data, set the coefficient at 0.9.

Then multiply 0.9*1.1*0.9 = whatever it is.
 
danminin:

How do you calculate the probability?

Let's say a bad nonfarm is released and you know from history that the chart goes down 70% of the time after such news.
You say "there's a 70% probability the chart should go down now!"
that's it! you use historical data!

or do you want to use technical analysis to calculate probability? give me an example of such a calculation.


There is a different situation at each moment in time. This is the basis on which we can build upon.

There are many indicators that we can analyse and draw conclusions from.

It is difficult to say unequivocally by calculation. I am working at the moment in this direction.

 
Uladzimir Izerski:

One does not preclude the other. But one must always assess the probabilities of the current moment.

The news background, important data and so on have to be taken into account however you want, and they may not fit with the historical results.

Well, I am only interested in history as a statistic and nothing more. The news background is probably also of interest to me - the effects are very contradictory and in the past few times I have been irritated by it.

What news background when there are several, from 3 to sometimes 10, trades a day in different directions? I do not look beyond one day on the clock history. I do not know where the market will go either. Where it will go, all right. As you have correctly noted - I am interested only in the current moment.

 
danminin:
What are the conditions? How do you calculate the current momentum?

Unless you set the coefficients.

For example, there are 3 news for the day
bad news, you set the odds at 0.9.
A good interest rate change. Put odds at 1.1.
A bad inflation data, set the coefficient at 0.9.

Then multiply 0.9*1.1*0.9 = whatever it is.

This is just the fundamental unit, and plus TA.

It is possible to squeeze out a good result.

 
Yuriy Asaulenko:

It's from the cycle - I haven't read it, but I know it.))

I never wrote or said that you can earn money by randomly wandering. Read carefully.)) And learn some maths, by the way, before you explain anything to anyone.))


Here is a quote:

Yuriy Asaulenko:

Finally, at least one has understood that the market is a random process. But if someone thinks that you can't play and win on a random process, that's their own business.



It's hard to explain the maths to people... ...considering they're mostly C's.)