which model(s) could i use?
I am trying to analyze which factors are more important when it comes to earning more 10 years after college.
So i think it is called causal modeling - i have data on different institutions and what their students have in terms of "average SAT scores, race, gender, study and how much chance there is that they earn 25.000 dollars per year 10 years after college"
i wish to figure out which of these attributes are the most important when it comes to having the highest % chance of earning (at least) 25.000 dollars per year 10 years after college.
i thought i could do a decision tree but all the examples i can find one have a binominal label (yes or no/Churn or Loyal) and my label ranges greatly from 25 % chance to 80 % chance.
any help is greatly appriciated.
Answers
Try the Feature Weighting operators.
hello @jera17ae - welcome to the community. Yes I would highly suggest looking at feature weighting as @Thomas_Ott suggests. Also I find using mod.rapidminer.com quite handy when needing to choose a model. Don't forget to look at the "Read Before Posting" etiquette (on right) where you can see how to post your XML process in a community thread.
Scott