Output Data Sets

OUT= Data Set

The OUT= data set contains all the variables in the input data set plus new variables that contain the principal component scores, residuals, and other computed values listed in Table 10.2.

The names of the score variables are formed by concatenating the value given by the PREFIX= option (or the default Prin, if PREFIX= is not specified) and the numbers 1, 2, ..., , where is the number of principal components in the model.

The names of the residual variables are formed by concatenating the value given by the RPREFIX= option (or the default R_, if RPREFIX= is not specified) and the names of the process variables used in the analysis. Residual variables are created only when the number of principal components in the model is less than the number of process measurement variables in the input data set.

Table 10.2 Computed Variables in the OUT= Data Set

Variable

Description

Prin1–Prin

Principal component scores

R_–R_

Residuals

_NOBS_

Number of observations used in the analysis

_SPE_

Squared prediction error (SPE)

_SPEMEAN_

Mean SPE for a given time value

_SPEVARI_

Variance of SPE for a given time value

_TSQUARE_

statistic computed from principal component scores

Note: The _SPEMEAN_ and _SPEVAR_ variables are included in the OUT= data set only when you specify the TIMEGROUP= option in the PROC MVPMODEL statement. These values are used by the MVPMONITOR procedure to construct control limits. See Organization of the Input Data Set and Example 11.1 for more information.

OUTLOADINGS= Data Set

The OUTLOADINGS= data set contains the eigenvalues of the correlation (or covariance) matrix, the loadings computed for the process variables, and the number of observations used in the analysis.

The number of observations in the OUTLOADINGS= data set is equal to the number of process variables, plus one observation that contains the eigenvalues. The variables saved in the OUTLOADINGS= data set are listed in Table 10.3.

Table 10.3 Variables in the OUTLOADINGS= Data Set

Variable

Description

_PC_

Principal component number; 0 for the observation that contains eigenvalues

_NOBS_

Number of observations used in the analysis

process variables

Principal component loadings for process variables


Note: This procedure is experimental.