Hearst index - page 17
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
In general, it does not count that way.
Please check that all three functions work correctly.
1. normal ISC
2. Total Least Squares
3. Adaptive one with weights which is exactly the reason of all the fuss.
My 'Useful Functions from KimIV', I have long tested it and checked. No errors.
The regular ISC take mine, 'Useful functions from KimIV' I tested it for a long time and checked it. No errors.
Just the regular one is the least of my worries :)
k[i] = 0.5/(0.5 + value*value/avgDev)
did you assume this yourself (and the whole further calculation) or can you share a link with a description?
k[i] = 0.5/(0.5 + value*value/avgDev)
did you assume this yourself (and the whole further calculation) or can you share the link with the description?
Yes, alas. You can substitute whatever you want.
The assumption is this -- The most common deviation will be between 0.5 and 1*avgDev.
Preference was given to 0.5 as it gives more insensitivity to outliers.
Please check the operation of all three functions.
Yes. Alas. You can use whatever you want.
The assumption is that the most common deviation will be between 0.5 and 1*avgDev.
Preference was given to 0.5 as it gives more insensitivity to outliers.
Please check all three functions.
I have it in a different way.
Post your calculation, then it will be clear what the difference is
This is not how I did it.
You can see the difference.
You got the same thing :) .
Multiply the numerator and denominator in your formula by Summ(k) and then look closely at my calculations :) .
{ //... // y = ax + b // counting a and b a = ekx*ekx - ekxx*ek;// Здесь считается ЗНАМЕНАТЕЛЬ // спецом чтобы можно было проверить ошибку деления на 0, если кому-то приспичит // второй круг посчитан a = (eky*ekx - ek*ekxy)/a;// Здесь считается числитель и делится на заранее посчитанный знаменатель b = (eky - a*ekx)/ek; //... }You got the same thing :) .
Multiply the numerator and denominator in your formula by Summ(k) and then look carefully at my calculations :) .
Or rather, multiply by minus -Summ(k)
We'll consider that we've conquered the issue :)
You got the same thing :) .
Multiply the numerator and denominator in your formula by Summ(k) and then look carefully at my calculations :) .
Listen, the result is a lot different than I thought it would be.
the new curve is more twitchy!!!!! instead of smooth :)
and also more amplitude.
and the curve is independent of the coefficient in k (0.5=1=2=...)
Look, it's a completely different result than I was expecting.
the new curve is more twitchy!!!!! instead of smooth :)
and also more amplitude.
and the curve is independent of the coefficient in k (0,5=1=2=...)
so i've done it right too. Told about it before - it jumps a lot ((.
I must have done the right thing too. Told you about it before - it jumps a lot ((
I just made a mistake in one place in the indicator.
>> the weighting does not work, the difference is in the thousandths.
Well, the fact that it bounces, that's true.