Details

Displayed Output

For each pass of the data (that is, for each BY group and for each pass required by the pattern of missing values for the dependent variables ), the GLMMOD procedure displays the definitions of the columns of the design matrix along with the following:

• the number of the column

• the name of the associated effect

• the values that the class variables take for this level of the effect

The design matrix itself is also displayed, along with the following:

• the observation number

• the dependent variable values

• the FREQ and WEIGHT values, if any

• the columns of the design matrix

Missing Values

If some variables have missing values for some observations, then PROC GLMMOD handles missing values in the same way as PROC GLM; see the Missing Values section on page 1837 in Chapter 32, The GLM Procedure, for further details.

OUTPARM= Data Set

An output data set containing information regarding the association between model effects and design matrix columns is created whenever you specify the OUTPARM= option in the PROC GLMMOD statement. The OUTPARM= data set contains an observation for each column of the design matrix with the following variables:

• a numeric variable, _ COLNUM_ , identifying the number of the column of the design matrix corresponding to this observation

• a character variable, EFFNAME , containing the name of the effect that generates the column of the design matrix corresponding to this observation

• the CLASS variables, with the values they have for the column corresponding to this observation, or blanks if they are not involved with the effect associated with this column

If there are BY-group variables or if the pattern of missing values for the dependent variables requires it, the single data set defines several design matrices. In this case, for each of these design matrices, the OUTPARM= data set also contains the following:

• the current values of the BY variables, if you specify a BY statement

• a numeric variable, _ YPASS_ , containing the current pass of the data, if the pattern of missing values for the dependent variables requires multiple passes

OUTDESIGN= Data Set

An output data set containing the design matrix is created whenever you specify the OUTDESIGN= option in the PROC GLMMOD statement. The OUTDESIGN= data set contains an observation for each observation in the DATA= data set, with the following variables:

• the dependent variables

• the FREQ variable, if any

• the WEIGHT variable, if any

• a variable for each column of the design matrix, with names COL1, COL2, and so forth

If there are BY-group variables or if the pattern of missing values for the dependent variables requires it, the single data set contains several design matrices. In this case, for each of these, the OUTDESIGN= data set also contains the following:

• the current values of the BY variables, if you specify a BY statement

• a numeric variable, _ YPASS_ , containing the current pass of the data, if the pattern of missing values for the dependent variables requires multiple passes

ODS Table Names

PROC GLMMOD assigns a name to each table it creates. You can use these names to reference the table when using the Output Delivery System (ODS) to select tables and create output data sets. These names are listed in the following table. For more information on ODS, see Chapter 14, Using the Output Delivery System.

Table 33.1: ODS Tables Produced in PROC GLMMOD

ODS Table Name	Description	Statement
ClassLevels	Table of class levels	CLASS statement
DependentInfo	Simultaneously analyzed dependent variables	default when there are multiple dependent variables
DesignPoints	Design matrix	default
NObs	Number of observations	default
Parameters	Parameters and associated column numbers	default