Averaging data with formula

[nurhazx]

Hi, 

I am trying to create monthly average data by ignoring the downtime condition (signal less than 60).

I am using formula :

Quote

 

$signal=$bt.remove($d)

$signal.aggregate(average(), $ar, startKey(), 500d).toDiscrete()

 

$d= downtime condition

$ar= monthly capsule

 

The value from formula is different from manual averaging as shown in pic below.

Could anyone help me to explain the difference?

 

Many thanks

[nurhazx]

The other month (Sep-Nov) value is same between formula and manual. Only that one. And this is happen in another signal.

[Patrick]

Hi Nurhazx - In general, the average displayed in the details pane will be dependent on the time range selected in your display pane.  i.e., in the above screenshot, 9/13/2023 to 3/14/2024. If you select a timerange by left-click and dragging a section of the trend, as you did in your screenshot, the average will representative for the selection - in the above, it will be 12/1/2023 to 12/13/2023.  The details pane average will also include data removed in your monthly aggregation since you are passing the `remove` operator into your monthly aggregation.  

The calculated monthly average will always be based on the start and end time of the monthly condition, regardless of the display pane time range.  It will of course also honor the "remove()" parameter.

One note regarding "remove()" - if you are removing periods shorter than your signal maximum interpolation, Seeq will interpolate (draw a line) across the "gap" of data removed.  If you, instead, want the removed section to be shown as a true gap regardless of interpolation settings, you can use the "within()" operator combined with "inverse()"

$signal=$bt.within($d.inverse())