The ODS OUTPUT statement will allow you to put any desired statistic into a SAS data set. Yes, you will need PROC APPEND as well. ... View more

PROC MEANS or PROC SUMMARY or PROC UNIVARIATE will provide you with the sum of a variable. The OUTPUT statement will provide the result in a SAS data set for you to use. ... View more

> I see that my program produces the files that I > need to export even though the error message is > there. Are the results outputted correct in this > case?? They are correct in my example. ... View more

I don't observe this. Could you provide more details? Or a copy of the log? What do you mean by "stops working"? What do you mean by "less no. of observations in a group"? ... View more

In what way do you mean "I'm trying to create a graph that has one variable (Y1) on the positive part of the Y-axis, and another variable (Y2) on the negative part of the Y-axis"?? More detail would help. ... View more

Before you call GPLOT, you should create a new variable that has Y1 and Y2 combined properly. ... View more

You would need to output other ODS tables that contain R-square and N ... View more

32 independent variables and no intercept? I certainly don't recommend doing that. Can't add variable number 32, no matter what it is? Perhaps you have run out of degrees of freedom?? The whole idea of having 32 independent variables and then doing some variable selection is usually doomed to fail. A better approach, in my opinion, is something like Partial Least Squares regression, where you don't have to drop variables, and you find linear combinations of independent variables that are predictive of Y. ... View more

PROC SCORE will let you use the estimated regression equation. ... View more

In most (all?) comparable situations, the results will be the same. GLM allows for categorical factors and interactions/polynomial terms. REG offers regression diagnostics. By the way, in answer to your question "which one should I use to create a model" ... you don't create a model in software ... the model is independent of software ... it can exist on a blackboard ... you have to specify the terms in the model. Software can estimate the parameters in your models. Message was edited by: Paige ... View more

Thanks for the explanation. ... View more

> so we all know that when we are working with mixed > effects models we have marginal and conditional > residuals. We do? This is a false statement. I don't know. Perhaps you could enlighten me. ... View more