SAS PROC MIXED: Model statement

[ Return Top SAS Home ]

The model statement specifies the fixed effect model. The outcome (dependent variable) is placed between the word model and the equal sign. The covariates (independent variables) are listed to the right of the equal sign with no punctuation between them. Two variable names with an asterisk in between signifies the interaction of those variables. In cosinor analysis the sin and cos variables created in the data step are used as covariates. An intercept is implicit in the model statement.

 model attrib = rx age sex cos24 sin24 cos12 sin12 cos8 sin8 cos6 sin6
    cos4p8 sin4p8
    rx*cos24 rx*sin24 rx*cos12 rx*sin12 rx*cos8 rx*sin8 
    rx*cos6 rx*sin6 rx*cos4p8 rx*sin4p8;

In this example the treatment indicator (rx) is interacted with each of the cos and sin terms to produce a model which can fit a different shape for each treatment.

The s option (at the end, after a slash) is required if you want to see the fixed effects coefficients (solutions).

We will use this at a later stage of the model fitting.

[ Return Top SAS Home ]