REPEATED Subcommand (MIXED command)

The REPEATED subcommand specifies the residual covariance matrix in the mixed-effects model. If no REPEATED subcommand is specified, the residual covariance matrix assumes the form of a scaled identity matrix with the scale being the usual residual variance.

  • Specify a list of variable names (of any type) connected by asterisks (repeated measure) following the REPEATED subcommand.
  • Distinct combinations of values of the variables are used simply to identify the repeated observations. Order of the values will determine the order of occurrence of the repeated observations. Therefore, the lowest values of the variables associate with the first repeated observation, and the highest values associate with the last repeated observation.
  • The VC covariance structure is obsolete in the REPEATED subcommand. If it is specified, it will be replaced with the DIAG covariance structure. An annotation will be made in the output to indicate this change.
  • The default covariance type for repeated effects is DIAG.
  • The following keywords, which are specific for the REPEATED subcommand, can be entered after the effects. Use the vertical bar (|) to precede the options.
    SUBJECT(varname*varname*…)
    Identify the subjects. Complete independence is assumed across subjects, thus producing a block-diagonal structure in the residual covariance matrix with identical blocks. The number of subjects is equal to the number of distinct combinations of values of the variables. A case will not be used if it contains a missing value on any of the subject variables.
    KRONECKER(varname*varname*…)
    The Kronecker product of two matrices can handle doubly repeated-measure data in which there are two repeated effects; one of them indicating the multivariate observation of target, and the other the time or the order of the data measured. The KRONECKER keyword should be used only when COVTYPE is defined as UN_AR1, UN_CS, or UN_UN.
    When both KRONECKER and REPEATED are in effect, REPEATED always comes first, defining the effect on the left-hand side of the Kronecker product structure. Both of them may or may not have common fields, but their values cannot be exactly the same (even when the values are not in the same order).
    COVTYPE(type)
    Covariance structure. Specify the covariance structure of the identical blocks for the residual covariance matrix (see Covariance Structure List (MIXED command)). The default structure for repeated effects is DIAG.
    SPCOORDS(varname...)
    Specifies the variables that define the spatial covariance coordinates. One or more numeric variables must be specified. SPCOORDS is required when SP_POWER, SP_EXPONENTIAL, SP_GAUSSIAN, SP_LINEAR, SP_LINEARLOG, or SP_SPHERICAL is specified for the COVTYPE keyword. It is ignored otherwise.
  • The SUBJECT keyword must be specified to identify the subjects in a repeated measurement analysis. The analysis will not be performed if this keyword is omitted.
  • The list of subject variables must contain all of the subject variables specified in all RANDOM subcommands.
  • Any variable used in the repeated measure list must not be used in the repeated subject specification.

Example

MIXED  SCORE BY CLASS
  /RANDOM = CLASS | SUBJECT(SCHOOL)
  /REPEATED = FLOOR | SUBJECT(SCHOOL*STUDENT).

However, the syntax in each of the following examples is invalid: 

MIXED  SCORE BY CLASS
  /RANDOM = CLASS | SUBJECT(SCHOOL)
  /REPEATED = FLOOR | SUBJECT(STUDENT).


MIXED  SCORE BY CLASS
  /RANDOM = CLASS | SUBJECT(SCHOOL*STUDENT)
  /REPEATED = FLOOR | SUBJECT(STUDENT).


MIXED  SCORE BY CLASS
  /RANDOM = CLASS | SUBJECT(SCHOOL)
  /REPEATED = STUDENT | SUBJECT(STUDENT*SCHOOL).
  • In the first two examples, the RANDOM subject list contains a variable not on the REPEATED subject list.
  • In the third example, the REPEATED subject list contains a variable on the REPEATED variable list.