Some guardrails for regression modeling

A simple checklist

After totally dumping on models yesterday, I thought it might be nice to do a little something to be part of the solution as opposed to the problem. So, here is a simple little checklist/flowchart to follow when building a regression model:

Asking these questions won’t necessarily mean you build a good model or that it won’t be used improperly or that things won’t go wrong. But, it’s better to satisfy these questions, which act as guardrails, than it is to ignore them. Happy modeling!

Links

Yesterday’s Post | Most Popular Posts | All Historical Posts | Main Site | Contact