Hi All,
I have a numerical continuous column and around 300 columns( character, nominal) of process variables. I want to understand the effect of different values caused by all these 300 columuns. I was thinking of profiler function so that I can see clearly which variable has the impact.
2 questions:
1. How to use the profiler function for non-numerical values? (I wasn't able to do that)
2. What is the best way to model the interdepedant factors of the variables and show a particular process has the highest probability of causing the issue or something like that?
Any comments/suggestions will be of great help.
Thank you.
A Decision Tree might be the way to go. You mentioned howmany columns you have, but nothing about how many rows of data. Howmany rows do you have?
