Hi, all
I am a new member to the community. I am assigned a task to perform recurrence analysis on our data. I never used JMP or did any recurrence analysis. So I will bring some stupid questions for sure.
The data we have are robots repair. We want to see repairs event pattern for the bots and I was told to use recurrence analysis (don't know why).
So we have around 15000 robots, I used the 'bot_id', 'accumulated_transactions', 'site', 'cost', 'bot_type' and 'component' columns. 'Site' includes different location where the robots work. 'cost' is the 0, 1 value ( 1 means a repair, and 0 means the last seen event of the bots). 'component' are different parts under repair. We want to see effect of sites, bot_type, and parts on mcf and repair pattern. I primarily use Log Linear NHPP model.
if I want to study the effect of site on mcf, should i put site as 'grouping' or 'cause' roles? If I want to find which site's bots deteriorate faster than others?, which site have more repair compared to others.
My understanding is that Grouping is for category factor that is exclusive to each other, like a bot cannot be in two location at the same time. So I should use Grouping. But ChatGPT suggested Cause. I don't know which is right.
Also I have a question on how to choose columns for scale and shape effect. For instance if I choose site as Cause as GPT suggested. what will be chosen for scale, shape effect. I felt that site could affect those two as well. And JMP allows me to choose site as well. My question is that Can I choose site as 'Cause' and 'scale' and 'shape' at the same time. It generates a result that I cannot explain(hundreds of MCF lines and parameter estimate tables of each site affecting all sites (including it self). So I guess that I cannot do that.
Then I put component (parts) as 'Cause', site as scale and shape factor. JMP generates a model for each component. After fitted in log linear NHPP, a parameter estimate table is generated for each cause. The issue is that the MCF curve is 12 components X 20 sites, generating hundreds of lines. I felt that I am doing the wrong thing here again.
Feels that if I use 'Cause' roles then I should not put anything as 'scale' or 'shape' effectors, or it will make the MCF curve too crowded.
Finally I decided to put site as 'Grouping' to compared different site, site as 'scale' and 'shape' effectors.
In a different analysis, I put components(parts) as 'Cause' to compare among parts (blank for 'scale' and 'shape'), and use local data filter to filter to each site to generate a MCF for a single site with different component(parts) cause. Does the design make sense now?
Sorry this is long.
