How about a calculation like:
if you would like to see this in use, please provide a packaged workbook with sample data that represents your situation.
Thanks Joe for your initial help. I've attached a twbx as suggested.
In the workbook, I'm trying to build a table like below, then produce a graph based on that.
Week of Tree# A Running Total Tree# B Running Total Cumulative Yield So Far Dec 30, 2012 164 176 340 Jan 6, 2013 183 183 366 Jan 13, 2013 192 196
Jan 20, 2013 201 202 403
last_in_partition.twbx.zip 24.3 KB
1 of 1 people found this helpful
How about the attached?
Date is used as the compute using, addressing, making Week partition, so the LAST() function returns 0 for the last date in each week.
Because you do not want to see each date's value, I used LAST() as a filter instead of an offset lookup.
To prevent Data Densification, the Date pill used for compute using is an integer instead of the date data type.
Then to connect the dots as a line, Week is a continuous, green, pill the date number pill is on the Path shelf.
This type of work is on the weird side of Tableau. There are other ways to accomplish this that may be a better fit for your situation.
Wow. Thanks for walking me through your solution. I've learned a new trick today
The chart is missing in [result jm] sheet, with the error '... Is Last requires a field that is missing. ...' . I tried adding date number pill in rows, and I got the points, no lines though. Is that right? Attached screenshot.
My next step is to combine tree #A and B result together, e.g. tree #A's result for date number 41296 of 201 should be added with B's 41297's 202 and plot that sum, while ignoring the other rows in the partition. Can tableau do this?
A little more background, this problem stems from my need to show a running total that starts calculation from the beginning of my data set vs tableau's running total that at the most covers what's visible on the graph. Perhaps, I'm approaching this whole thing wrong