Elite indicators :) - page 404

 

52 week high/low

Can anyone create one indicator to display 52 week high/low as horizontal line in Daily chart?

Thanks in advance

 

...

Here is a kind of generalized high-low indicator. Default parameters are set as you asked for, but you can choose any time frame and any period for that

regards

Mladen

clc4x:
Can anyone create one indicator to display 52 week high/low as horizontal line in Daily chart? Thanks in advance
Files:
 

Mladen

Good afternoon Mladen,

Trust this finds you well rested and comfortable enough to give me an honest opinion about an indicator I asked Kerris to build me years ago. I do know how you feel about working on older indicators, but the problem with this indicator is that it refreshes every hour when in MTF mode which I use to step it up to M30 for M5 charts, etc.

Would you be kind enough and check it out for me to see if anything could be changed to stabilise it plse? I do not mind it resetting after a day's trading as I understand the mechanism involved in MTF coding.

Thanking you in advance.

Files:
rsi-m5.mq4  6 kb
 
mladen:
Been fooling with custom rsi function (since built in rsi has some problems) and while doing it tried it on various things. Came up with this too. Not a new idea but the implementation is (starting from not using built in rsi)


It is a rsi of averages. Averages possible are for the averages indicator and you can choose :

.

0 - Simple moving average

1 - Exponential moving average

2 - Smoothed moving average

3- Linear weighted moving average

4 - Linear regression value (or LSMA)

5 - Triangular moving average

6 - Sine weighted moving average

7 - Volume weighted moving average

8 - Hull moving average

9 - Non lag moving average

10 - Parabolic weighted moving average

Here is an example of an rsi of a triangular moving average :

Hi Mladen

Divergence

 

...

TepeFinans

Just to let you (and everybody else) be in current : decided to make a completely new divergences "module" (easily patchable to any indicator) that will have a "tolerance" (it will not require strict peaks at the same bar (which I think is the biggest problem of the current way of finding divergences) but it will compare peaks "somewhere in the neighborhood") as well as arbitrary "depth" of divergences (for example comparing peak 1 with peak 3, peak 4 and so on for depths different then 1) but it will take some time to brush it out (at least a couple of days of testing and cleaning up the code). As soon as it is finished, will post it here ...

regards

Mladen

TepeFinans:
Hi Mladen Divergence
 
mladen:
TepeFinans

Just to let you (and everybody else) be in current : decided to make a completely new divergences "module" (easily patchable to any indicator) that will have a "tolerance" (it will not require strict peaks at the same bar (which I think is the biggest problem of the current way of finding divergences) but it will compare peaks "somewhere in the neighborhood") as well as arbitrary "depth" of divergences (for example comparing peak 1 with peak 3, peak 4 and so on for depths different then 1) but it will take some time to brush it out (at least a couple of days of testing and cleaning up the code). As soon as it is finished, will post it here ...

regards

Mladen

hey mldaen. always been a fan of you work!

why not just compare the most recent peak and the peak previous to it? that way, it would make the divergence "the most current pattern found" if the peaks don't make a divergence, then no patter found? pretty simple and it would be a lot less code maybe?

you can get hidden divergences as well using it.

it would read like this in the indicator (lets use MACD for example)

present candle has a MACD value above 0 at say .0187 then the next candle and the candle before it have lower MACD values. take that and compare the very last time MACD was above 0 and the values made a peak then fell. take that and compare it to the highs (or lows for hidden divergence) then make an alert.

just some thoughts.

 

hazelj80

I agree with you. What I am trying to do is to implement some "tolerance" (for example if peaks are at an "acceptable distance" too and not only on at exactly the same bar - which was the cause for previous divergence code to omit some obvious divergences sometimes - simple because the peaks were not at the same bar). I am trying to find some way of automatically "deciding" if the peaks are "close enough" logic. Also, I would like to see what are we going to get if we compare 1st and 3rd, 1st and 4th, and so on ... peaks in order to see multiple (so in a way "confirmed") divergence I think that the confirmation will add (at least some) value to divergence detection.

PS: thanks for the thoughts, I always thought that those are the best things we can share (code is just a result of thoughts after all )

regards

Mladen

hazelj80:
hey mldaen. always been a fan of you work!

why not just compare the most recent peak and the peak previous to it? that way, it would make the divergence "the most current pattern found" if the peaks don't make a divergence, then no patter found? pretty simple and it would be a lot less code maybe?

you can get hidden divergences as well using it.

it would read like this in the indicator (lets use MACD for example)

present candle has a MACD value above 0 at say .0187 then the next candle and the candle before it have lower MACD values. take that and compare the very last time MACD was above 0 and the values made a peak then fell. take that and compare it to the highs (or lows for hidden divergence) then make an alert.

just some thoughts.
 
mladen:
hazelj80

I agree with you. What I am trying to do is to implement some "tolerance" (for example if peaks are at an "acceptable distance" too and not only on at exactly the same bar - which was the cause for previous divergence code to omit some obvious divergences sometimes - simple because the peaks were not at the same bar). I am trying to find some way of automatically "deciding" if the peaks are "close enough" logic. Also, I would like to see what are we going to get if we compare 1st and 3rd, 1st and 4th, and so on ... peaks in order to see multiple (so in a way "confirmed") divergence I think that the confirmation will add (at least some) value to divergence detection.

PS: thanks for the thoughts, I always thought that those are the best things we can share (code is just a result of thoughts after all )

regards

Mladen

excellent! look forward to it.

P.S. have you had any thoughts of a better Elliot Wave indicator than what's out there now without the zigzags? that would be a tough one but I think it's doable.

 

corrected average MTF

Hi there,

I did tried but i can't make it... ...

Mladen or MrTools could you please make corrected average https://c.mql5.com/forextsd/forum/144/corrected_average.mq4 MTF?

Really, really appreciated....

Thanks ya in advance and

Best Regards

 

...

brax64

Here you go Added the rest of the averages that were accumulated in the mean time (so now there are 17 types of ma in it - MaMode from 0 to 16)

regards

Mladen

brax64:
Hi there,

I did tried but i can't make it... ...

Mladen or MrTools could you please make corrected average https://c.mql5.com/forextsd/forum/144/corrected_average.mq4 MTF?

Really, really appreciated....

Thanks ya in advance and

Best Regards
Reason: