Do books help you in trading? - page 5
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
Oleg, I'm very friendly to you, I really am. Maybe I spoke out of turn, I'm sorry. I don't like to tease.
I understand what you're saying, I know what frequencies and filters are, I just don't see their application here. I assume you have a different perspective on it, and get it "cooked" somehow. If you're interested in telling me that, now or later, I'd love to hear from you.
I do not like the average one for its lag. I prefer it to the minima and maxima. If an extremum is exceeded it is visible instantly, not after some time. I did not decompose price movement into its frequency components. I simply do not see any possibility of profit there. That is why I do not take the average one seriously. That's the way it is.
It happens. Bygones.
Your preference (lows and highs) is essentially filtering too.
But it would be interesting to compare results of your indicator and my indicator.
.
Your preference (lows and highs) -- it's essentially filtering too.
Yes, you're right, between those two values the system is asleep and filters out a decent chunk of information. I did a comparison with a normal sliding average. Here are the results:
-registering a new extremum instantly, as I said, at the average through the averaging period,
-registration of a new trend also occurs earlier than the crossing of averages of different periods,
-One more thing: the divergence - it actually means when a new maximum hasn't "broken" the previous one, but has come very close, sometimes even point to point. You can see it at once as well.
As for your picture, it looks fine. I don't know the algorithm, but it looks good
Regarding your picture, yes in general it looks accurate. I don't know the algorithm, but it looks good.
The picture is not indicative at all, the standard makd looks good at this scale. A lot of noise in real time hides inside the candle and is not visible on the history due to very short period. It is necessary to see the indicator in terms of low timeframe, all its problems will appear there.
Yes, you're right, between those two values the system is asleep and filters out a decent chunk of information. I did a comparison with a normal sliding average. Here are the results:
-registering a new extremum instantly, as I said, at the average through the averaging period,
-registration of a new trend also occurs earlier than the crossing of averages of different periods,
-One more thing: the divergence - it actually means when a new maximum hasn't "broken" the previous one, but it is close to it, sometimes even to a point. You can see it at once as well.
As for your picture, it looks fine. I do not know the algorithm, but it looks good.
And show me a picture of what your indicator shows in this area, for clarity.
The picture is not illustrative at all, the standard mcd looks good at this scale. A lot of noise in real time is hidden inside the candle and is not visible on the history due to very short period. The indicator should be converted to a lower timeframe, all its problems will appear there.
What is noise in your understanding?
This is it:"You must see the indicator in terms of the low timeframe, all its problems will appear there" is nonsense due to misunderstanding.
This is the youngest timeframe, the most :
.
What problems are you talking about?
What is your definition of noise ?
And this:"You have to see the indicator in terms of low TF and all its problems will appear there" is nonsense due to misunderstanding.
This is the youngest timeframe:
.
What problems are we talking about?
Recalculation to a lower timeframe means that you take H4, for example, instead of Daily, and multiply the values of the operating parameters by 6. Preferably more history with different price behaviour, and in a normal vertical scale.
Recalculation to a lower timeframe means that instead of Daily you take for example H4 and multiply the values of the operating parameters by 6. Preferably more history with different price behaviour, and in a normal vertical scale.
Oh, that's it... I see... you can also get on your head...
On D1 I have a bunch of bars, here on H1:
I tried to pick up similar averages so that approximately the crossover points coincided with the trend registration points. Outwardly it seems to be less, but in the flat the averages make a lot of crossovers and there is no new trend registration because the extrema are not exceeded.
Here is a flat section with the same parameters:
That is, there is an obvious upward trend in this section (highs are increasing), while the averages are floundering.
Here is a flat section with the same parameters:
That is, this section is clearly trending upwards (highs are increasing), while the averages are floundering.
So this is the flat section. There is no trend in this area. However, it depends on the interpretation.