Requests & Ideas - page 66

 

john was that a live forward test?

if it aint broke dont fix it

Odysseus

 
odysseus11:
john was that a live forward test?

if it aint broke dont fix it

Odysseus

No, it's just historical test. Does live forward test mean running this EA in realtime with real or demo account? I have been running the EA with this setting on my vps for a week. It works well until now, but i think it's just too early to come to a conclution.

 

brax64

Try out this one

It is a modification that I was working on a long, long time ago with some extras (showind mid levels, showing "baby levels", showing bars when a line is "touched" alerts, and so on ...) As far as I remember it was working OK
regards

Mladen

brax64:
Hi everybody,

I've the Murray Math indicator attached that seems to have a refreshing problem; I mean the lines become shorter (seems that the right end of the lines didn't update the position each bar) and also (but I'm not sure about that) seems that didn't update the calculation when the Highs/Lows of reference changes with the time.

Now I need to change time frame once in a while to force the indi to recalculate.

Also the labels below the lines have the origin position in the middle, so I had to add spaces on the text to keep them aligned with the left end of the lines; there is a way to set the origin of the labels at the left side?

Could anybody have a look in the code and fixing it?

Your help is really appreciated!

Regards
Files:
 

New Pitchfork?

Mladen,

Please take a look at the attachments

Screen is a normal andrews pitchfork.The Median Line goes trough the 50% retracement in price and time,of points 2 and 3 of the pitchfork,as usual

Screen2 is andrews pitchfork+2 additional "Median Lines",one goes trough 38.2%,in price and time,the other goes trough 61.8%.

2 questions :

1-Will it be possible to modify andrews pitchfork to automatically do what I manually did in screen2 picture?Showing the 3 median lines,and,if possible not showing the Fibonacci retracements ?

2-In case it is possible,will it be possible to add 2 more Median Lines for the 23.6% and 76.4/78.6% retracements?

3-Sometimes,I prefer to work with octaves instead of Fibos(0.125,0.250,0.375,0.500,0.675,0.750,0.875),that means we would need 6 additional median lines to the 0.5 already included in andrews pitchfork...Will it be possible or we will be restricted by buffer limits?

Thanks in advance for your consideration and Regards.

S

Files:
screen.gif  65 kb
screen2.gif  70 kb
 

Mladen, have a big request for you:

I have been testing the PolyFit Scalper EA, and one thing I notice is that you have developed two VERY interesting alternate versions of Polyfit bands V2.3

One of them is the PA (phase accumulation) polyfit bands 2.3 and the other is an alb polyfit bands 2.3

My request is that I would like to "insert" these variants for use/testing in the Polyfit scalper EA, but I have no idea what indivcator settings for these indicator variants correspond or "equal" the current suggested regular polyfit bands for use in the EA.

The settings for "normal" polyfit bands in the EA v1.59 nohedge are:

price 0

length 150

degree 1

presmooth 1

MAmode 0

sigma 2.8

can u "translate" for me so that I know what starting settings to use on both the PA polyfit bands indi and alb polyfit bands indi to "equal" these above setting for "regular" polyfit bands 2.3? (their settings are completely different and I dont understand the translation). Thanks a million!

Odysseus

 

Odysseus

Unfortunately there is no translation from a regular to adaptive ones. The adaptive ones are changing periods all the time depending on market conditions and thus one can never know what is the actual period calculated so you can not tell that for example period nn is xx cycles in pa, for example. I usually set the default parameters in the adaptive indicators to resemble to values of the non-adaptive ones with default values, but that is just resemblance

The rule is more or less the following : the longer the periods the less chance is that it will look like the adaptive one since the changes in periods in the adaptive one are going to make big differences values in volatile periods. It is a rulle valid for anu adaptive indicator (even jurik MA which is also an adaptive indicator)

regards

Mladen

odysseus11:
Mladen, have a big request for you:

I have been testing the PolyFit Scalper EA, and one thing I notice is that you have developed two VERY interesting alternate versions of Polyfit bands V2.3

One of them is the PA (phase accumulation) polyfit bands 2.3 and the other is an alb polyfit bands 2.3

My request is that I would like to "insert" these variants for use/testing in the Polyfit scalper EA, but I have no idea what indivcator settings for these indicator variants correspond or "equal" the current suggested regular polyfit bands for use in the EA.

The settings for "normal" polyfit bands in the EA v1.59 nohedge are:

price 0

length 150

degree 1

presmooth 1

MAmode 0

sigma 2.8

can u "translate" for me so that I know what starting settings to use on both the PA polyfit bands indi and alb polyfit bands indi to "equal" these above setting for "regular" polyfit bands 2.3? (their settings are completely different and I dont understand the translation). Thanks a million!

Odysseus
 

OK thanks - then could u tell me what settings for PA bands (filter/cycles) and alb bands (swing count) would be good defaults to approximate a static period of 150 (which is used in the "normal" v 3.2 bands in the EA)? I can go from there.

thanks!

odysseus

 

odysseus

The easiest way is to take the "underlying" indicator and see when does it give "expected" periods

Here is an example : this is a phase accumulation indicator itself with cycles set to 5 (so how many bars does it take to add up phases of every singe bar to sum 5 whole cycles) on a 1 hour EURUSD chart - then it reaches 120 periods, but also see how much the periods are varying. You have to experiment and test different time frames and symbols since it really depends on data and not some absolute math formula
regards

Mladen

odysseus11:
OK thanks - then could u tell me what settings for PA bands (filter/cycles) and alb bands (swing count) would be good defaults to approximate a static period of 150 (which is used in the "normal" v 3.2 bands in the EA)? I can go from there.

thanks!

odysseus
Files:
pa_5_cycles.gif  19 kb
 

ah I see - thanks!

Odysseus

 

simba

Just an update : I am doing something regarding the last request (the Nonlinear trend line) and it is currently at this state (these are just 2 quick working examples with 5 control points, but points are going to be freely definable and one will be able to move them around at wish) :

There are still some things to do (things that need to be checked and rechecked) and some things how the indicator should "remember" its states when closing metatrader (that probably will be the most tricky thing to do - metatrader does not make a difference of indicator removing and closing the metatrader itself)

__________________________________

Anyway, am tired a bit now (must grab some sleep) and will continue tomorrow, but the real test will be only on Monday (no way to test some things in back test - this one is really a specific one)

So, I hope it will be finished soon and then we shall have one more tool that is at least very rare at metatrader (and it seems on any other platform too - frankly I have been looking around and I haven't found anything usable on tradestation, ninja trader, metatrader (one interpolation that is posted for metatrader (this one : https://www.mql5.com/ru/code/8566 ) is anything but usable for this purpose, and frankly, I fail to see its usability at all in the form it is)) not just because of this way of using splines (since this is a spline ) but I am eager of testing spline interpolation in some sort of multi time frame or with some peak finding stuff (who knows what can we get )

Good night to all now

Files:
spline.gif  28 kb
spline_1.gif  29 kb