Econometrics: let's discuss the CU balance sheet. - page 9
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
Well, why can't we use the profit derivative due to IR variability? MT is just the hospital average... You can't use IR as a determining factor in determining the robustness of a system.
yes, variation in mo is allowed, and mo is the hospital average. But if Mo varies within a small range (i.e. Mo distribution itself is statistical:)), then the hospital average will be representative and the residuals will be normally distributed.
P.S. I myself do not use econometrics and do not analyse balances. I use simpler trader's indices, and I also use visual equity analysis.
I will publish the answer as promised above - we need normality of residuals only for reliable interval estimates (to calculate the width of the confidence interval) - an important procedure for applied problems. But if interval estimates are not needed - we can build a regression for any distribution of both the observed value in the sample and the residuals.
So whether the residuals are distributed normally, abnormally, with thick tails, with thin tails, without tails - all the same..................................................
Some mix of different things, and the purpose is lost.
We traders have no purpose in assessing the correctness of interval estimates. It's an auxiliary tool, necessary to determine the credibility of the resulting figures.
If we built the regression and the residue is normal - analysis is complete and we are lucky with the TS.
This has never happened in my short practice. The residue is not normal and not even stationary. The regression figures cannot be trusted. What to do next? If you put the word "balance" in my statements, what does it mean that the test results cannot be trusted. That's what it's about. It's the target, not the correctness of the regression model.
Yes, mo variability is allowed, and mo is a hospital average. But if Mo varies in a small range (i.e., Mo distribution itself is stable:)), then the average is indicative and the residuals will be distributed normally.
P.S. I myself do not use econometrics and do not analyse balances. I use simpler trader's indices, and I also use visual equity analysis.
no-no, you use "secret" methods which you "are not hired to describe here".
No, no, you use "secret" methods that you "are not hired to describe here".
yeah, they're simpler than that. jealous? :)
Some confusion of different things and the goal is lost.
We traders have no purpose in assessing the correctness of the interval estimates. It is an auxiliary tool, necessary to determine the credibility of the obtained figures.
If we built the regression and the residue is normal, the analysis is over and we are happy with the TS.
This has never happened in my short practice. The residue is not normal and not even stationary. The regression figures cannot be trusted. What to do next? If you put the word "balance" in my statements, what does it mean that the test results cannot be trusted. That's what it's about. It's the target, not the correctness of the regression model.
Then we do not understand each other - normality of the residual is not necessary if interval estimates are not needed. A model without normality of residuals will be correct and adequate with a certain accuracy if the series is stationary.
Why would you need such a model? Profit forecasting - yes, if the series is stationary. Although in Al.ar there are a couple of PAMM charts, where the balance curve was steadily growing and then rapidly fell down (collapse) - the classical non-stationarity.
Yes, with them and the simpler ones. Jealous? :)
of course! I love all sorts of "secret knowledge" and "world conspiracy" worshippers.
Yes, mo variation is allowed, and mo is the hospital average. But if mo varies within a small range (i.e. mo distribution itself is statistically significant:)), then the hospital average will be representative and the residuals will be normally distributed
P.S. I myself do not use econometrics and do not analyse balances. I use simpler trader's indices, and I also use visual equity analysis.
Residuals should not be distributed normally.
Normal distribution is an idealization.
I have long observed a widespread fascination with this idealization...
Then we do not understand each other - normality of the residual is not necessary if interval estimates are not needed. A model without normality of residuals will be correct and adequate with a certain accuracy if the series is stationary.
What does stationary mean? How is it determined?
The residuals do not have to be normally distributed at all.
A normal distribution is an idealisation.
I have long observed a widespread fascination with this idealisation...
Well, give me an example of a "good one" from your point of view where the residuals are not normally distributed
So if profitability of TS is supposed to be constant, i.e. mo=const, why do we need some complicated detrending instead of just subtracting the linear trend from equity? I.e. trend model y=kx where k=mo, x-transactions, y-equity
How complicated? Instead of "trend" symbols wrote "HP".
But there are more serious considerations. The analytical straight line smoothing formula (more accurate than detrending) is very much dependent on the sample size. Let's take the EURUSD sample since 2000. Let's isolate the trend as a straight line. almost a horizontal straight line, but with deviations of about 2500 pips! This is exactly what the machine writes - the average hospital temperature. But if we take any filter we will get variance of tens of pips. Since we are not trading on time intervals of 10 years, we can do with a straight line when smoothing out 50-100 observations. But some estimations require more observations. I always apply a filter to avoid getting into details. Purely practical consideration.