Skip to main content
Announcements
See what Drew Clarke has to say about the Qlik Talend Cloud launch! READ THE BLOG
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Calculating % change - Year over year

Hello,

I am trying to calculate the % change of allowed dollars PMPM between years. I am currently using the following formula:

(above( sum([AMT_ALLOWED])/ sum({1}MM_UNITS))/( sum([AMT_ALLOWED])/ sum({1}MM_UNITS))) -1

Note - i need the "/ sum({1}MM_UNITS))" to stay in the expression to get an accurate PMPM. Because if someone wants to restrict by a claim parameter, I still need my underlying population to stay constant (because i need members for PMPM, not claimants).

However, if I make any selection to restrict my data, I encounter the following problem (Say I only want to see 'Owners' and 'PR Individual')--

     Before I made the "%Change Allowed" field, anything that was excluded from my selection would collapse. Now everything remains and is populated with nulls/zeros. Any suggestions on how to change my formula above to get the excluded fields to collapse again? Or perhaps a better way to calculate % change between years as shown above?


Thank you.

1 Solution

Accepted Solutions
Not applicable
Author

Thanks Gysbert - that didn't work for this scenario, but i found a way to simplify a few of my other expressions and reference them.

Changed my expression for 'MMs' to:

sum({1<BENCHMARKS=P(BENCHMARKS),INCURRED_YEAR=P(INCURRED_YEAR),INCURRED_YEAR_AND_MONTH=P(INCURRED_YEAR_AND_MONTH)>} MM_UNITS)

'Total Alwd PMPM' to: sum([AMT_ALLOWED])/[MM's]

'%Change Alwd' to: ([Total Alwd PMPM])/ below([Total Alwd PMPM])-1

This way i was able to avoid using the {1} in my '% change' column altogether, by using my set analysis on the front end in my MM's expression.

View solution in original post

2 Replies
Gysbert_Wassenaar

Perhaps the problem is that you're dividing by zero for the other benchmark values because of the {1}. Maybe this works: alt((above( sum([AMT_ALLOWED])/ sum({1}MM_UNITS))/( sum([AMT_ALLOWED])/ sum({1}MM_UNITS))) -1,0)


talk is cheap, supply exceeds demand
Not applicable
Author

Thanks Gysbert - that didn't work for this scenario, but i found a way to simplify a few of my other expressions and reference them.

Changed my expression for 'MMs' to:

sum({1<BENCHMARKS=P(BENCHMARKS),INCURRED_YEAR=P(INCURRED_YEAR),INCURRED_YEAR_AND_MONTH=P(INCURRED_YEAR_AND_MONTH)>} MM_UNITS)

'Total Alwd PMPM' to: sum([AMT_ALLOWED])/[MM's]

'%Change Alwd' to: ([Total Alwd PMPM])/ below([Total Alwd PMPM])-1

This way i was able to avoid using the {1} in my '% change' column altogether, by using my set analysis on the front end in my MM's expression.