Exploring, predicting and reporting with SAS Visual Analytics and SAS Visual Statistics

How to create a Pareto Chart

Occasional Contributor
Posts: 5

How to create a Pareto Chart

Hello SAS-Community,

would you have any idea how to create a Pareto Chart in SAS Visual Analytics. I'm interested to create 2 Pareto Chart.

The first one using a category and the frequency.

The second one using a category and a measure.

Those are the type of charts I can use in our SAS VA version.

Thanks a lot for your support

Best Regards,


Regular Contributor
Posts: 191

Re: How to create a Pareto Chart


you must have already selected a non numberic variable, so the histogram  (Pareto) is greyed out.  (Historgam can not deal with a non numberic value).

Please start with an enpty screen and directly select the Historgrom before you select anay data fields.

If you did not yet select any variable, I have no idea why the Histogram icon is greyed out.  (and thats the one you need....Smiley Wink)

Good luck. Smiley Happy

Occasional Contributor
Posts: 5

Re: How to create a Pareto Chart

Posted in reply to PeterWijers

Hi Peter,

creating the histogramm is fine for me, but how can I create a cumulative measure and cumulative percentage measure based on the used measure in the graph like e.g. frequency or kg and having a category on the x-axis, so that I can build the pareto?

Thanks for your help

Regular Contributor
Posts: 191

Re: How to create a Pareto Chart

Hi Marcus,

At first my mistake to mix up capability chart with Pareto. Sorry for that. 

A very simple presentation would be to select Bar-Chart and drag fill the category. Now the bar chart will automatically sort in large to small freq order.

This looks like a Pareto but has no cumulative line.

Seems to be you have VA 6.3 installed, I am running on 6.2  (To be updated soon)

Other option would be to use a Bar-Line report and create calculated fields on group level for this.   (I will check later if this also can be done in 6.2

I know that VA 6.3 has new functionality here, but I do not have it available yet.

Thats it for now......waiting on the upgrade to 6.3.


SAS Employee
Posts: 11

Re: How to create a Pareto Chart

Posted in reply to PeterWijers

Hi Markus,

From the UI screenshot I can tell you are using the VA Explorer. Peter's suggestion of using a Bar-Line would require you to switch to the VA Designer.

The missing piece to get a Pareto chart is to calculate the cumulative totals. As of the current release, VA can't do this calculation for generic categories; only for dates. This is something we are likely to add in a future version though.

Occasional Contributor
Posts: 13

Re: How to create a Pareto Chart

[ Edited ]

For the benefit of new users like me, here is how I created my first Pareto Chart using SASUE Sudio:


I have used example given in the book - Statistical Methods for Quality Improvement by Hitoshi Kume. Table 3.2, Page 20.


Step1: Created XL file with the following columns: Type of Defects|Number of Defects|Percentage of Defects|Cumulative Percentage


Step 2: Sort the table in ascending order of Number of defects (As shown in the picture)

Step 3: Import the xl to SASUE. From the table select the column names for graph as shown:

As you see, I have chosen, Type_of_Defect| Number_of_Defects|Cumulative_Percentage column names.



Step 4: Go to Tasks and Utilities

Step 5: Select Bar-Line Chart

Step 6: Select Category variable as Type_of_Defect field

Step 7: Select Bar Response variable as Number_of_Defects field

Step 8: Select Line Response variable as Cumulative_Percentage


Run the program. With little tweaking of chart options, we get Pareto chart output.



For your use, data file enclosed. 


Since I am also a new user of SAS University, I welcome comments, suggestions for improvements from experienced forum members.


I will require some help in changing "Number of defects" axis, where I want to show total number of defects. Here is the output from XL chart:



Hope this helps!


Kind regards,



Ask a Question
Discussion stats
  • 5 replies
  • 4 in conversation