[SOLVED] Seeking to model patterns in claim payment history
I have a series of payment transactions in a standard columnar report, including fields such as:
claim #
transaction date
transaction type
payment amount
My tasks are 3-fold (using Info Assist only):
1 - Count the # of transactions per claim and come up with an average # for the entire record set.
2 - Calculate average # of days between payment dates.
3 - Visualize the payment timing and size pattern by cause of loss, indicating a confidence interval where 95% of the time, payments fall into the pattern shown.
Who is feeling especially brilliant today?
Please keep it simple as I am but a newbie Info Assist business user.
Many thanks in advance!This message has been edited. Last edited by: Tamra,
WebFOCUS 8.10.5 Windows 10, HTML
July 11, 2016, 11:35 AM
Tamra
HI Gavin B,
While we wait for others to reply, here are a few suggestions to point you in the right direction.
1) Count the # of transactions per claim and come up with an average # for the entire record set.
You can apply "aggregation Functions" such as COUNT and AVERAGE to a field to. Click on the field / right mouse click/ MORE / Aggregation Function
- to get the average of the entire data set that may require a bit more coding - but don't get alarmed - this would be set up within a "compute" or you may need to create a HOLD file
2) Calculate average # of days between payment dates.
In a COMPUTE you can get the number of days between 2 dates by using a FUNCTION - select the stylized fx icon to pull up the functions DATEDIF will provide the difference in days between 2 dates.
The AVERAGE function with a Filter on the date should answer this one or you may need to create a HOLD file with the a count of the number of days and then another report on the HOLD to give the average for a sort value.
3) May require a HOLD to get the data into the format that you want then a report with the visualization on this data.