Is there any way to aggregate an average without outliers?

[ofir_y]

I need a way to create a transform that will aggregate the average of a field but without the outliers (let's say all values that falls between 10%-90% percentiles). for example if I have the following values:
[1,2,3,4,5,6,7,8,9,10]

it will calculate the average of 2-9

[Hendrik_Muhs]

Unfortunately there is no out of the box solution for this.

What I can think of:

You could in addition to you existing group_by further group by histogram. In the aggregation you need to calculate an average for that bucket and you need the count of documents (value_count on one of the group_by fields).

The transform would create a document for every histogram bucket. In a 2nd pass you can query the transform dest index, using a range query to filter out the outliers and aggregate using a weighted average aggregations, this is where you need the count as weight.

The other idea: filter the outliers already in the transform or use a filter aggregation in the transform with a avg child aggregation.

To get the left and right cut off value you can use a percentiles aggregation.

To sum it up, I do not see a solution that can be easily implemented but both ideas require extra work and additional queries or at least 2 transforms.