Bayesian regression - Has anyone made an EA using this algorithm? - page 16
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
1. The likelihood is maximal at : the long formulas go on. We can say that we get the minimum value of the mean square of the residuals, or we can say that we have maximised the likelihood.
2. there may be something you don't understand. the coefficient b1 is what? The mathematical expectation of the sample values of coefficient b1, which is t-distributed in the absence of knowledge of the parameters of coefficient b1 on the general population. Linear regression (ordinary least squares) gives an estimate of E(b) and sigma(b), the standard error of coefficient b1. What you see in the output of the model is all these estimates. Then there is an estimate of how significantly different E(b) is from 0, the t-statistic and the associated probability.
3. I can't say anything about trends. Symmetry is important - fact. Sigma on the residuals is also important. The coefficient of kurtosis is also important.
4. I've been reading a lot about regression recently, so what I've written above I understand. I report to my customers on regression results and have to understand something. Although I prefer non-parametric methods.
1. Actually regression is Bayesian, in case you forgot. So you need to maximise the value of the Bayesian probability formula.
2. This is complete nonsense of the mind of an inflamed mathematician.
3. So you're doing something stuck in t-distributions, normality, estimators, sigmas... with no understanding of what you're doing or why, or what it is in the first place. Then why and for what are you continuously fighting the normality of the distribution here.
4. "Reporting on regression results" is something!
1. Actually the regression is Bayesian, in case you've forgotten. So you need to maximise the value of the Bayesian probability formula.
2. This is complete nonsense of an inflamed mathematical mind.
3. So you're doing something stuck in t-distributions, normality, estimators, sigmas... with no understanding of what you're doing or why, or what it is in the first place. Then why and for what are you continuously fighting the normality of the distribution here.
4. "Reporting on regression results" is something!
I think you should refer to the classic methods manual: http://www.intuit.ru/studies/courses/1153/318/info
I'm not offended by the boorishness.
I think you should refer to the classic methods manual: http://www.intuit.ru/studies/courses/1153/318/info
I'm not offended by the rudeness.
How about you don't tell me where to go and I don't tell you where to go?
How about you don't tell me where to go and I don't tell you where to go?
Which ward will you be from?
There is nothing to take there seriously. In fact, the problem is solved at the level of a term paper of a 4th year student of some department related to automation.
Dimitri, look at the absolute coincidence of the sums of the actual and estimated population values, can any method achieve this coincidence and is this fact random?
Even after that (18) did not surprise you? True, such a miraculous result is achieved by a special case of (18), when n=1 and its graph is more monotonous:
1. Dimitri, look at the absolute coincidence of the sums of the actual and estimated population values, can any method achieve this coincidence and is this fact random?
2. Even after that (18) did not surprise you? It is true that a special case of (18) achieves such a miraculous result, with n=1 and its graph is more monotonic:
There was an interesting article about some scientist inventing some rays in the "Tekhnika Zhurnika" magazine in 1980. This scientist decided to show his experiments to an outside observer. The experiment was conducted in the dark, and this outside observer took the sensor and unscrewed it, and after that the arrow of the device was just dangling from the noise. Nevertheless, the scientist saw how the instrument reacted to his beams.
1. No surprise there. Your regression is a regression with the same exponent as shown in the graph, so it will coincide with it. The fact is not random but fabricated, not all processes develop on such graphs. The same coincidence can be obtained with a polynomial regression, even better.
2. Not at all. The regression method was developed long ago and clearly. It has long been known that you can take any function, either with one parameter, or with two, or with any other number of parameters. By some mathematical calculations, you can get formulas. So your discovery is a standard problem for a 3rd or 4th year math student (or maybe 2nd year). In other words, there is a curvuline function, there is a coincidence check function, and the coefficients of curvuline construction are calculated using well-known methods (taking the derivative and solving a system of equations).
Yousufkhodja Sultonov, where can I download the latest version of your indicator?
Here is the working version, the latest one - I will post it soon
Thanks - I'll try to assess the effectiveness in my PBX.
Turns out I already have it - so it's an old version...
It keeps redrawing on past bars, which makes it impossible to properly assess its effectiveness, why is this done?
Judging by this part of the code
SetIndexBuffer(i, Buy); SetIndexLabel(i, "Buy"); SetIndexStyle(i, DRAW_ARROW); SetIndexArrow(i, 233); i++;
SetIndexBuffer(i, Sell); SetIndexLabel(i, "Sell"); SetIndexStyle(i, DRAW_ARROW); SetIndexArrow(i, 234); i++;
Arrows should be drawn, but they are not.
SELL and BUY buffers are present, but there is no information in them. Wouldn't it be better to put there a +1/-1 histogram for buying and selling?