JMP variability chart

#1

Hi,

I am a JMP software user and I have just joined the community. I am trying to see if I can replace JMP with Knime. I frequently use a chart in JMP called the Variability Chart which looks like this:

image

It is basically a box plot or a scatter plot which separates and plots data based on groups of variables used as x-axis. The example above shows only a two-level grouping for the x-axis but you can have as many as you wish. It is a very handy chart for depicting large sets of data. I was wondering if it is possible to generate something like that through an existing Knime node or through one of its extensions (R, Python, etc.). Any insight is appreciated.

Best Regards,
Sherwin

0 Likes

#2

Hi Sherwin,

we are happy to hear that you consider to switch from JMP to KNIME.

This particular plot is a bit tricky. There is no easy way to make a box plot with multi-level groups. For a single group split you can use

This kind of visualisation looks indeed useful so I will make a ticket in our internal system to add it to TODO list.

Looking around, it seems that such plot is also not available in either python or R out-of-the-box. You can do it for 2 levels showing second level as a color. See python example and R example. But I was not able to find tools to display such plots for more than 2 levels (other than showing multiple individual plots side-by-side).

Cheers,
Mischa

2 Likes

#3

Thanks Mischa for the time you took to respond.
It is a bit surprising to see that apparently only JMP can produce such charts given how brilliant and useful they are. I hope to see it added to the Knime’s visualization tools in near future.

Best Regards,
Sherwin

0 Likes