Calculating rolling average without a date

[Anonymous]

Hello

I'm a little stuck and hope you can help?  I need to calculate a rolling average of ITIssues over a time series but the time series does not contain a date, just period numbers.  My example data is:

Period	Period Number	Year	Index	ITIssues
2017/18 P1	P1	2017/18	0	11
2017/18 P2	P2	2017/18	1	15
2017/18 P3	P3	2017/18	2	10
2017/18 P4	P4	2017/18	3	15
2017/18 P5	P5	2017/18	4	15
2017/18 P6	P6	2017/18	5	11
2017/18 P7	P7	2017/18	6	9
2017/18 P8	P8	2017/18	7	6
2017/18 P9	P9	2017/18	8	7
2017/18 P10	P10	2017/18	9	8
2017/18 P11	P11	2017/18	10	7
2017/18 P12	P12	2017/18	11	9
2018/19 P1	P1	2018/19	12	7
2018/19 P2	P2	2018/19	13	12
2018/19 P3	P3	2018/19	14	8
2018/19 P4	P4	2018/19	15	11
2018/19 P5	P5	2018/19	16	6
2018/19 P6	P6	2018/19	17	9
2018/19 P7	P7	2018/19	18	7
2018/19 P8	P8	2018/19	19	10
2018/19 P9	P9	2018/19	20	14
2018/19 P10	P10	2018/19	21	9
2018/19 P11	P11	2018/19	22	14
2018/19 P12	P12	2018/19	23	7

The period numbers do not correspond to equal date periods eg period 1 in 2017/18 is not the same date as period 1 in 2018/19.  I need to see a rollling average over the periods.

Many thanks

Jo.

[Greg_Deckler · Accepted solution]

I would create a new calculated column that is essentially the numeric equivalent of your Period like:

201818

201819

...

Then you could use the same basic concept of a rolling average without the date. You should be able to do something like:

Column = SUBSTITUTE([Column1],"/","") * 1

This may also help:

See if my Time Intelligence the Hard Way provides a different way of accomplishing what you are going for.

https://community.powerbi.com/t5/Quick-Measures-Gallery/Time-Intelligence-quot-The-Hard-Way-quot-TIT...