This post was originally published by my partner, Aureus Analytics, a specialist in Insurance Analytics. It is a very interesting illustration of how one would choose a statistical model. I thought you may find it just as interesting and it may help those who are new to analytics. 

Views: 25222


You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by D A B N Amarasekera on June 29, 2015 at 6:20pm

Very nice. But if you can add at the beginning why and when modelling is done, it is good.

Comment by D A B N Amarasekera on June 29, 2015 at 6:12pm

Very nice summary.  But if you can add at the beginning when and why modelling is done 

Comment by Jay Jarman on June 28, 2015 at 4:44pm

I teach a data/text mining course to undergrads and graduate students and am always looking for a new way to help students figure out how to select an algorithm from which to build a model.  Maybe I'm missing something here but the flow chart doesn't tell how to choose a statistical model.  It just regurgitates my selections at the top part of the tree.  

Comment by Tim Boomer on June 26, 2015 at 11:04am

I like this. Has anyone extended the approach to include tree-based methods, SVM, and the like?

Comment by Baguinebie Bazongo on June 26, 2015 at 7:42am

Very nice

I would add "Research question" as the first step to check for

Because research question leads to depend variables and independ variables


© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service