I am trying to create an individual and moving range chart by date on the x-axis. I know it is possible given the following article.
However, I am unable to "drop" the dates on the x-axis. Everytime I try, JMP assumes that I am placing it in "subgroup" and
converts to an Xbar and R chart. Does anyone know where I need to drop my dates in the control chart builder
to change the x-axis? Any help would be much appreciated!
I can't seem to replicate that problem with simulated data however this can happen if have duplicated dates. Can you increase the detail on the date data?
I'm unsure of what you mean by duplicated dates. I do have the same date for multiple data points but I am confused as to how this solves the problem. The type for my test date is continuous. I tried changing the type to ordinal but the graph still changed to an Xbar and R chart. Thank you for responding.
When you have the same date for multiple data points it will result in an Xbar/R chart. Can you change the date to include the time of day so each date is an individual rather than a "group". I did this and it worked fine
Alternatively you can just use the Quality&Process>Control Chart>IR platform and it will behave as you like. The Control Chart Builder has AI built in so when it sees duplicate dates it wants to create a subgroup.
Thank you again for responding so quickly. I have tried that. Unfortunately, my manager wants the scale to be in "log" which is not possible with the IR platform.
Now you totally lost me. What scale to be in log the Y response or X axis? If you change the date to a time stamp that includes day and time of day then each X axis will be unique.
Okay cool and now I see what you were doing. In CCB you are able to change the scale to log scale as an option yet you can not do that in the Control Chart IR platform. I was given some help on an option for you. Add a new column next to your date column and indicate it as sample number (on a given day). CCB will allow you to add date and also the sample number which will give you a nested unique X and then you should be able to still use CCB and log scale.