Signal Bar

Optuma Scripting
1

Hi,
I am looking for a Function that identifies a larger than average bar (i.e. the range from Low to High). The idea is to find Signal Bars as described by MS Jenkins, where a final bar of the move turns the price trend. That bar is a Signal Bar.

There is also another aspect to this where the price close has to be at an extreme of the move. However this would be easy to script.

Cheers, Lester

2

Hi,

Before a script could be built you would need to define the rules of what constitutes larger than average.

Ie, do you take the average range for the last 20 bars and look for something 50% bigger than that range? Once we have the rules, the script can be built around those.

3

[postquote quote=67505]

I’m curious about Jenkins’ signals. Where did you find it in his publications?

4

Hi Matthew,
Thank you for your reply.

I do not have a firm idea on what constitutes a larger than normal bar range. I was hoping there was a function that I could use that would have variable parameters to adjust for different bar range group averages and percent differences. If such a function existed then it may be possible to home in on whether Signal Bars could be reliably identified.

I did have a look through the script dictionary but only found the Congestion Index that seemed to be vaguely related. The BarType closest to what I am looking for is an Outside Bar, but that is no good because it depends on the preceding bar. Even if that could be made to work it would only be a rough method as no parameters could be added.

Cheers, Lester

5

Hi David,
I found the term in MS Jenkins 2010 Training Video. I think I have also seen it in one or more of his books as well.

Cheers, Lester

6

Hi,

We don’t have a predefined script function, the parameters would need to be specified as a rule set, and each rule programmed into the script manually.

7

Hi Lester,

Further to my earlier reply, here is an example script of one way larger than average daily ranges can be identified. It may not be exactly what you are after, but will give you an idea to start from, and build from there.

//Set Bar Range to a % movement value
V1 = 100 * abs(Low() - High())/High();
// Find yearly bar movement average
V2 = MA(V1, BARS=252, CALC=Close) ;
//Find monthly bar movement average
V3 = MA(V1, BARS=20, CALC=Close) ;
// Find Bars where the % movement is 2.5 times higher than the yearly and 3 times higher than the monthly average
V1 >= V2 * 2.5 and V1 >= V3 * 3

Here’s an example of bars that would pass the above criteria…

Ex11

8

Hi Matthew,
Thank you for your reply, I shall see if I can develop it further.

In the mean time I put together a script (which at this stage does not perform well). I used the logic of using the difference of a Moving Average High and subtracted a Moving Average Low to get an average range. Since normal Moving Averages are out of synch with the trend I used a Hull MA. Because there is no option for selecting OHLC I made both HIGH() and LOW() variables in the HMA.

My script is below.

Cheers, Lester

v1 = HIGH(); v2 = HMA(v1, BARS=10); v3 = LOW(); v4 = HMA(v3, BARS=10); Average Range = v2 -v4; Daily Bar = v1 - v3; Daily Bar > Average Range * 1.5