Direkt zum Hauptbereich

Tableau Table Calculation Function: WINDOW - Functions



The functions which begin with „WINDOW_...“ are also common used in Tableau.

Remember!
The “WINDOW_” function stays for the offset in data set, so-called WINDOW. It can look like this:
1) You can see the “WINDOW” clearly because of separation line between rows:



2) You limit the “Window” by giving the information about the first and the last row number. In this case, you give Tableau the information about the data offset.


Let's have a look at the example with WINDOW_SUM
I created a sample with data from Superstore.




I would like to have a total sum of Sales in every row. In order to do this I created a calculation field:
WINDOW_SUM(SUM([Sales]), FIRST(), LAST())

With this formula I said to Tableau:
“Hey Tableau, calculate the total sum of sales from the first till the last row in the data set”

And this is the result:



Tableau wrote the result (total sum of Sales) in every row.

As another option you can cumulate the result in each row and beginn from any row you wish. For this you need this formula:

WINDOW_SUM(sum([Sales]),FIRST()+3,0)




Try out follow combination and see what happens with data:

  • WINDOW_SUM(sum([Sales]),FIRST()-3,0)
  • WINDOW_SUM(sum([Sales]),FIRST()-3,5)
  • WINDOW_SUM(sum([Sales]),FIRST()-3,LAST()-2)
  • WINDOW_SUM(sum([Sales]),FIRST()+3,LAST()-2)
If I don’t use the functions FIRST() and LAST(), Tableau runs the calculation from below to top. Look at the example below. For this example I used this formula: 

WINDOW_SUM(sum([Sales]),3,5)




You can run different operations and calculate the average, minimum, maximum, standard deviation etc.. It depends on, what are you going to use at the end of the function WINDOW_.....

The point is that you understand the logic of this function. I hope, I could help you understand this logic with this blog.

You can find here the blog article as an example of using of this function:
Here you can find explanation and examples to other windows functions

Best regards



Labels:

Kommentare

Kommentar veröffentlichen

Beliebte Posts aus diesem Blog

How create a calender with tableau?

I posted a calendar to tableau public, which I have done with Tableau 9.3. In my first blog I would like to share with you the steps for creating a calendar. The data sources and calendar you can find here. There are different reasons to use a calendar in tableau e.g. You can create your own calendar with an individual format and a lot of appointments. So it will help you make planning better.  You can also analyze things with calendar, if you drop&down a relevant dimension in order to show successful or not successful days.  To create such things is just a training and will help you use dimensions&figures on another way. And thi is my approach: I created a spreadsheet in excel with three columns: Date  Type of holiday  Holiday The column "Type of holiday" is necessary in order to make differences between private and official holidays. With the spell "C" I marked all private birthdays and with the spell "F" all offici...
8 Kommentare
Mehr anzeigen

I want to understand RegEx! (Part2)

It is already two weeks ago since my last blog post. This two weeks I spent in Scotland. It is a very beautiful country with unforgettable landscaps. However, it is not the topic I want to talk about. In this blog I will continue the topic about the RegEx and its using in Tableau. Tableau offers you four different uses cases of RegEx: REGEXP_EXTRACT If you have data like this: aspirin_20_mg_oral_tablet and you would like to extract onliy 20_mg. What I do is I check my RegEx on this web site: http://regexr.com/ If it is ok, put this RegEx in Tableau function: REGEXP_EXTRACT([0-9][0-9]+[_]+[a-z]+) and extract all "20_mg" REGEXP_EXTRACT_NTH The same like the function before but here you can limit your extract to Nth index. Image you have data like this: aspirin_20565435 and you want to extract only "aspirin_20". So your formula should be: REGEXP_EXTRACT_NTH([a-z]+[0-9], 2) REGEXP_MATCH It is very easy one: Using REGEXP_MATCH we can find if ...
Kommentar veröffentlichen
Mehr anzeigen