Hello, I'm using a formula that I was kindly given once on this forum. It has served me well - I have cut n paste it many times and it has always worked till now.
It's an array formula (CSE).
What is says is filter down to within a date range (C7 and D7 are dates) and then filter down to match a text condition (B10 is a text string). Then count the unique records in the range Q10 - Q74,865 on the allregs tab.
By using auto filter and "remove duplicates" feature I know that I should have 3,413 unique observations but my formula returns 314.
I appreciate this may be a longshot without seeing the data - if anyone can see anything wrong with the formula do tell me!
Thanks for any feedback,
Doug.
It's an array formula (CSE).
SUM(IF(FREQUENCY(IF('all regs'!$C$10:$C$74865>=summary!$C$7,IF('all regs'!$C$10:$C$74865>=summary!$D$7,IF('all regs'!$D$10:$D$74865=summary!$B10,MATCH('all regs'!$Q$10:$Q$74865,'all regs'!$Q$10:$Q$74865,0)))),ROW('all regs'!$Q$10:$Q$74865)-ROW('all regs'!$Q$10)+1),1))
What is says is filter down to within a date range (C7 and D7 are dates) and then filter down to match a text condition (B10 is a text string). Then count the unique records in the range Q10 - Q74,865 on the allregs tab.
By using auto filter and "remove duplicates" feature I know that I should have 3,413 unique observations but my formula returns 314.
I appreciate this may be a longshot without seeing the data - if anyone can see anything wrong with the formula do tell me!
Thanks for any feedback,
Doug.