Model Is misspecified how should I proceed

[Ciaran Deegan]

Hi All,

running the ovtest for my model returns a P-value of 0.0000. Adding squared or cubed terms for any of my variables does not move the needle here.

How should I proceed - is there a problem with my data? Or will I need to look at non-linear models?

Best,

Ciaran

[Carlo Lazzaro]

Ciaran:
welcome to this forum.
Please see the FAQ on how to post more effectively.
That said,asuming that, as per -estat ovtest-, you went -regress-, it may be that your regressand needs to be logged (but see also -glm-) and/or you missed one (or more) interactions.
Obviously, sharing what you typed and what Stata gave you back via CODE delimiters would help enormously interested listers to reply more positively. Thanks.

[Ciaran Deegan]

Thank you Carlo,

I will run the model again to include interaction terms. Also, I will repost with more detail, having read the FAQ's.

Best,

Ciaran