Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
Anonymous
Not applicable

How to Calculate 3 month moving average in set analysis

Hi All,

I Need help to get a 3 month Moving Average in set analysis, below is the source data for reference, i am using the below expression to get the ConversionRate%

count(DISTINCT{$<  %ProcessingType={'Disbursement'},PROCESS,[Rework Group]= ,[Month Num]>} CASE_NUM)/

count(DISTINCT{$<  %ProcessingType={'Creation'},PROCESS,[Rework Group]=,[Month Num] >} CASE_NUM)

Month#Disb#NewConversionRate%
Jan-16                1,799              2,04988%
Feb-16                1,187              1,97760%
Mar-16                1,587              1,67395%
Apr-16                1,102              1,22490%
May-16                1,074              1,74262%
Jun-16                1,091              1,92257%
Jul-16                1,066              1,30981%
Aug-16                1,197              1,77767%
Sep-16                    201                  46144%

Now i need the 3 month Moving average as below (i,e For March value=March+Feb+Jan)

Month#Disb (3month)#New(3Month)ConversionRate%
Jan-16                1,799              2,04988%
Feb-16                2,986              4,02674%
Mar-16                4,573              5,69980%
Apr-16                3,876              4,87480%
May-16                3,763              4,63981%
Jun-16                3,267              4,88867%
Jul-16                3,231              4,97365%
Aug-16                3,354              5,00867%
Sep-16                2,464              3,54769%

I tried using "ACCUMULATE 3 Steps Back in Expression" it applied only for Disb & New but not for the Conersion rate%, Can anyone help me to write a set analysis to achive the 3 months moving conversionRate% as in the 2nd table.

Thanks In Advance.

Arjin.

4 Replies
arulsettu
Master III
Master III

maybe like this

count(DISTINCT{$<  %ProcessingType={'Disbursement'},PROCESS,[Rework Group]= ,[Month Num],date={">=$(=MonthStart(Max(date), -3))<=$(=Max(date))"}>} CASE_NUM)/

count(DISTINCT{$<  %ProcessingType={'Creation'},PROCESS,[Rework Group]=,[Month Num],,date={">=$(=MonthStart(Max(date), -3))<=$(=Max(date))"} >} CASE_NUM)

date is your date field

Anil_Babu_Samineni

Arul, How about this?? Please correct me if i am wrong, Because he might looking current stage Date for last 3 months

count({$<  %ProcessingType={'Disbursement'},PROCESS,[Rework Group]= ,[Month Num],date={">=$(=AddMonths(Max(date), -3))<=$(=Max(date))"}>} CASE_NUM)/

count({$<  %ProcessingType={'Creation'},PROCESS,[Rework Group]=,[Month Num],,date={">=$(=AddMonths(Max(date), -3))<=$(=Max(date))"} >} CASE_NUM)

Best Anil, When applicable please mark the correct/appropriate replies as "solution" (you can mark up to 3 "solutions". Please LIKE threads if the provided solution is helpful
Anonymous
Not applicable
Author

Thanks for your time Arul & Anil,

i used the below expression and it worked fine for me.

sum( aggr( rangesum( above( Count( {$<Month,%ProcessingType={'Disbursement'},PROCESS,[Rework Group]=,[Month Num]>}  CASE_NUM),0,3) ),Month))/

sum( aggr( rangesum( above( Count( {$<Month,%ProcessingType={'Creation'},PROCESS,[Rework Group]=,[Month Num]>}  CASE_NUM),0,3) ),Month))

Thanks to Elifs post Accumulative Sums.

Cheers,

Arjin.

sunny_talwar

Try this:

RangeSum(Above(Count(DISTINCT{$<%ProcessingType ={'Disbursement'}, PROCESS, [Rework Group], [Month Num]>} CASE_NUM), 0, 3))

/

RangeSum(Above(Count(DISTINCT{$<%ProcessingType = {'Creation'}, PROCESS, [Rework Group], [Month Num]>} CASE_NUM), 0, 3))