topic Log tranformed dependent: PROC Mixed / NLMIXED? in Statistical Procedures

Log tranformed dependent: PROC Mixed / NLMIXED?

mragaa — Sun, 05 Nov 2017 01:31:07 GMT

Hi everyone,

I'm sorry in advance if this is a dumb question. Say I'm using panel data to model log(Speed) as the response and Turbo as a binary predictor (0= No, 1= Yes). My question:

1- Is Proc Mixed usable here, or do I have to use PROC NLMIXED?

2- Would the results be valid and interpretable as they are in cross-sectional data (i.e. we get x% in speed for Turbo)?

SAS Version: 9.4

Thank you.

Re: Log tranformed dependent: PROC Mixed / NLMIXED?

Rick_SAS — Mon, 06 Nov 2017 15:00:43 GMT

1. You can use PROC MIXED and other linear regression procedures if you first use the DATA step to define logSpeed = log(speed). Then use logSpeed as the response variable.

2. I am not sure what you are asking, but perhaps it has to do with the interpretation of the parameter estimates for Turbo. If you model

log(Speed) = Intercept + A*Turbo then

Speed = exp(Intercept)*exp(A*Turbo)

Thus when Turbo=0, the predicted speed is exp(Intercept).

When Turbo=1, the predicted speed is exp(Intercept)*exp(A).

So the model predicts that using Turbo results in MULTIPLYING the speed by exp(A).

Re: Log tranformed dependent: PROC Mixed / NLMIXED?

mragaa — Fri, 10 Nov 2017 13:56:26 GMT

Thanks a lot.