topic Re: Demonstrating Pareto's Law with JMP graph in Discussions

Demonstrating Pareto's Law with JMP graph

PearlyQuek — Sun, 11 Jun 2023 11:15:52 GMT

Hi everyone,

I will like to create a graph to show that 80% of the revenue comes from the top 20% of the customers in a store. My current data columns are:

1. CustomerID (around 4000+ data points)

2. Revenue

Would there be a way to build this graph in JMP? I've tried the Pareto Plot but have not been too successful.

Thank you! Cheers.

Re: Demonstrating Pareto's Law with JMP graph

Phil_Kay — Wed, 07 Jul 2021 08:56:22 GMT

Hi,

Would this kind of graph help?

Example data table attached.

Regards,

Phil

Re: Demonstrating Pareto's Law with JMP graph

PearlyQuek — Wed, 07 Jul 2021 09:30:08 GMT

Thank you Phil!

Yes it helps greatly. I was wondering if you could share how you did the ordering, as well as how you obtained the cumulative sales line? In addition, would it be possible for the cumulative sales to be in % instead?

I opened the Graph Builder script and could only see this:

Re: Demonstrating Pareto's Law with JMP graph

Phil_Kay — Thu, 08 Jul 2021 12:48:06 GMT

I am not sure why you graph looks so different when you run the Graph Builder script. I think it is partly due to preferences that you have set. But it should show the same cumulative sales curve.

(view in My Videos)

There are a few clicks involved so I thought it would be best to record a short video demo.

Some key points to achieve the graph:

We order the Company axis by Sales by dragging Sales to the drop zone above centre of the x-axis.
We select order descending after right-click on the x-axis labels
We add Sales on the y-axis then we add it again (drop in the zone to the centre-right of the y-axis)
We right-click on the y-axis labels to move one of the instances of the Sales variable to the right
We change the statistic of the right y-axis to Cumulative Sum
We right-click on the white space in the plot to change between bar and point plots for the variables

Are you able to recreate the graph now?

Cheers,

Phil