Using P() and Weighting/Proportion in Set Analysis

kmstephenson — Wed, 18 Jul 2018 17:12:33 GMT

Hi All,

I'd like to use weighting/proportions in set analysis with the P() function to weight the variable used in the possible function.

EXPRESSION - Readmit Rate % for all patients, per coverage selections:

Sum({1<Coverage=P(Coverage)>}[Numerator])/Sum({1<Coverage=P(Coverage)>}[Denominator])

Coverage Type:

For example, based on my selections, let's say that the distribution of Coverage Type is the following:

Current Selections Coverage Type Distribution:

A 200

D 0

N 1000

E 100

The P() works to exclude D from the calculation - however, what if I want to weight/use proportions based on the total?

Total Coverage Type Distribution:

A 500

D 2000

N 10000

E 200

Proportion based on Current Selections/Total:

A 200/500=40%

D 0

N 1000/10000=10%

E 100/200=50%

Is there any way to account for these proportions in the expression?

Thanks!

Re: Using P() and Weighting/Proportion in Set Analysis

sunny_talwar — Wed, 18 Jul 2018 17:15:19 GMT

I am not entirely sure I get, but may be you need this

Sum([Numerator])/Sum({1}[Denominator])