test for linearity in linear regression

noetsi

Fortran must die
#1
The partial slope coefficients are one way to test for this but I wondered if there was a formal test for linearity in linear regression (for a given predictor).
 

hlsmith

Not a robit
#2
As I said before, I am a fan of throwing it into a general additive model (GAM) and seeing if the relationship looks linear or not, you can control for other variables as well when doing this.
 

noetsi

Fortran must die
#3
If I knew the general linear model I might do that although I don't know any formal tests for non-linearity in that.
 

hlsmith

Not a robit
#4
Much like many parts of statistics, a visual review is sufficient. Though, GAM procedures will output , I believe, a polynomial - AKA number of knots. If visually the relationship appears linear and the values outputted is close to one, fitting a linear model is sufficient. This is solely based on my personal thoughts and experience, but seems reasonable.
 

noetsi

Fortran must die
#5
I have decided, after rereading Fox to use the partial regression plots. He, and others recommend it, I just have concerns about my judgement which is why I look for formal test.