Decision-making error - page 8

 
Yuriy Khrustalov:
Is a forecast a trading strategy?

A forecast differs from a strategy only in that there are no rules about how to accompany the position. Although, of course, it depends on what kind of forecast.

My point does not change here. And in a forecast, all points must be strictly justified.

 
George Merts:

A forecast differs from a strategy only in that there are no rules about how to accompany the position. Although, of course, it depends on what kind of forecast.

My thought does not change here. And all points in the forecast should be strictly justified.

You need to decompose the function of the predicted dependence into a Fourier series?

I'm not reading the script for those who can't read the script in black and white ("Variation of trend"))).

 
Yuriy Khrustalov:

You need to decompose the function of the predicted dependence into a Fourier series?

For those for whom the inscription in Russian in white on black ("Variation of the trend") is not readable)))

Well, it's OK, if this expansion has shown good results on the history - that's what you should do. For those interested, just write it that way - "approximation by Fourier series expansion".

Immediately a question to you - did you succeed in doing it (using decomposition)? The thing is that I was very interested in this area about a year ago, but I got convinced that such a prediction does not work. There are more errors than guesses. How about you?

 
George Merts:

So it's OK, if this decomposition has shown good results on your history, then you should do it that way. For those who are interested - just write so - "approximation by Fourier series decomposition".

Immediately a question to you - did you succeed in doing it (using decomposition)? The thing is that I was very interested in this area about a year ago, but I got convinced that such a prediction does not work. There are more errors than guesses. How about you?

The longer the period of history the closer to 50x50 , as with any prediction more variables equalize the probability ))))
 
My conclusion: For a successful TS you only need 1% which gives you a 50/50 advantage
 
Yuriy Khrustalov:
My conclusion: for a successful TS you only need 1% which gives you a 50/50 advantage.

I don't get it. Explain.

What do you mean by "need 1%"? You mean the winning percentage is 50.5 and the losing percentage 49.5? And what is the ratio of TP/SL? If the ratio is 2 or more, it is a good TS.

 
George Merts:

I don't get it. Explain.

What do you mean by "need 1%"? You mean a winning percentage of 50.5 and a losing percentage of 49.5? And what is the ratio of TP/SL? If the ratio is 2 or more, it is a good TS.

You got it all right. But just 1% is not enough for a good strategy, it turns out just a random number generator.)
 
Again the same as always, the move went where I planned, but after the stop was blown. I don't think any percentage can do anything about it. (
 
Yuriy Khrustalov:
Again the same as always, the move went where I planned, but after the stop was blown. I don't think any percentage can do anything about it. (
Maybe the stop is too close?
 
Yuriy Khrustalov:
Again, everything is as usual, the movement has gone where I planned, but after the stop was hit. I think here no percentage is able to do anything. (

What do you mean by "as always, but after the stop is off" ? Where do you put the SL? On the history - how often did it go out? Why do you have the SL there and not lower or higher? If you "always" correctly indicate the direction of movement, but often hit the SL - maybe the SL should be set further away? Checking the history, what does it say ?