The MIXED Procedure

ODS Table Names

Each table created by PROC MIXED has a name associated with it, and you must use this name to reference the table when using ODS statements. These names are listed in Table 59.26.

Table 59.26: ODS Tables Produced by PROC MIXED

Table Name

Description

Required Statement / Option

AccRates

Acceptance rates for posterior sampling

PRIOR

AsyCorr

Asymptotic correlation matrix of
covariance parameters

PROC MIXED ASYCORR

AsyCov

Asymptotic covariance matrix of
covariance parameters

PROC MIXED ASYCOV

Base

Base densities used for posterior sampling

PRIOR

Bound

Computed bound for posterior rejection sampling

PRIOR

CholG

Cholesky root of the estimated $\mb {G}$ matrix

RANDOM / GC

CholR

Cholesky root of blocks of the estimated $\mb {R}$ matrix

REPEATED / RC

CholV

Cholesky root of blocks of the estimated $\mb {V}$ matrix

RANDOM / VC

ClassLevels

Level information from the CLASS statement

Default output

Coef

$\mb {L}$ matrix coefficients

E option in MODEL,
CONTRAST, ESTIMATE,
or LSMEANS

Contrasts

Results from the CONTRAST
statements

CONTRAST

ConvergenceStatus

Convergence status

Default

CorrB

Approximate correlation matrix of fixed-effects parameter estimates

MODEL / CORRB

CovB

Approximate covariance matrix of fixed-effects parameter estimates

MODEL / COVB

CovParms

Estimated covariance parameters

Default output

Diffs

Differences of LS-means

LSMEANS / DIFF (or PDIFF)

Dimensions

Dimensions of the model

Default output

Estimates

Results from ESTIMATE statements

ESTIMATE

FitStatistics

Fit statistics

Default

G

Estimated $\mb {G}$ matrix

RANDOM / G

GCorr

Correlation matrix from the
estimated $\mb {G}$ matrix

RANDOM / GCORR

HLM1

Type 1 Hotelling-Lawley-McKeon tests of fixed effects

MODEL / HTYPE=1 and
REPEATED / HLM TYPE=UN

HLM2

Type 2 Hotelling-Lawley-McKeon tests of fixed effects

MODEL / HTYPE=2 and
REPEATED / HLM TYPE=UN

HLM3

Type 3 Hotelling-Lawley-McKeon tests of fixed effects

REPEATED / HLM TYPE=UN

HLPS1

Type 1 Hotelling-Lawley-Pillai-
Samson tests of fixed effects

MODEL / HTYPE=1 and
REPEATED / HLPS TYPE=UN

HLPS2

Type 2 Hotelling-Lawley-Pillai-
Samson tests of fixed effects

MODEL / HTYPE=1 and
REPEATED / HLPS TYPE=UN

HLPS3

Type 3 Hotelling-Lawley-Pillai-
Samson tests of fixed effects

REPEATED / HLPS TYPE=UN

Influence

Influence diagnostics

MODEL / INFLUENCE

InfoCrit

Information criteria

PROC MIXED IC

InvCholG

Inverse Cholesky root of the
estimated $\mb {G}$ matrix

RANDOM / GCI

InvCholR

Inverse Cholesky root of blocks of the estimated $\mb {R}$ matrix

REPEATED / RCI

InvCholV

Inverse Cholesky root of blocks of the estimated $\mb {V}$ matrix

RANDOM / VCI

InvCovB

Inverse of approximate covariance matrix of fixed-effects parameter estimates

MODEL / COVBI

InvG

Inverse of the estimated $\mb {G}$
matrix

RANDOM / GI

InvR

Inverse of blocks of the estimated $\mb {R}$ matrix

REPEATED / RI

InvV

Inverse of blocks of the estimated $\mb {V}$ matrix

RANDOM / VI

IterHistory

Iteration history

Default output

LComponents

Single-degree-of-freedom estimates that correspond to rows of the $\mb {L}$ matrix for fixed effects

MODEL / LCOMPONENTS

LRT

Likelihood ratio test

Default output

LSMeans

LS-means

LSMEANS

MMEq

Mixed model equations

PROC MIXED MMEQ

MMEqSol

Mixed model equations solution

PROC MIXED MMEQSOL

ModelInfo

Model information

Default output

NObs

Number of observations read and used

Default output

ParmSearch

Parameter search values

PARMS

Posterior

Posterior sampling information

PRIOR

R

Blocks of the estimated $\mb {R}$ matrix

REPEATED / R

RCorr

Correlation matrix from blocks of the estimated $\mb {R}$ matrix

REPEATED / RCORR

Search

Posterior density search table

PRIOR / PSEARCH

Slices

Tests of LS-means slices

LSMEANS / SLICE=

SolutionF

Fixed-effects solution vector

MODEL / S

SolutionR

Random-effects solution vector

RANDOM / S

Tests1

Type 1 tests of fixed effects

MODEL / HTYPE=1

Tests2

Type 2 tests of fixed effects

MODEL / HTYPE=2

Tests3

Type 3 tests of fixed effects

Default output

Type1

Type 1 analysis of variance

PROC MIXED METHOD=TYPE1

Type2

Type 2 analysis of variance

PROC MIXED METHOD=TYPE2

Type3

Type 3 analysis of variance

PROC MIXED METHOD=TYPE3

Trans

Transformation of covariance parameters

PRIOR / PTRANS

V

Blocks of the estimated $\mb {V}$ matrix

RANDOM / V

VCorr

Correlation matrix from blocks of the estimated $\mb {V}$ matrix

RANDOM / VCORR


In Table 59.26, Coef refers to multiple tables produced by the E, E1, E2, or E3 option in the MODEL statement and the E option in the CONTRAST, ESTIMATE, and LSMEANS statements. You can create one large data set of these tables with a statement similar to the following:

ods output Coef=c;

To create separate data sets, use the following statement:

ods output Coef(match_all)=c;

Here the resulting data sets are named C, C1, C2, etc. The same principles apply to data sets created from the R, CholR, InvCholR, RCorr, InvR, V, CholV, InvCholV, VCorr, and InvV tables.

In Table 59.26, the following changes have occurred from SAS 6. The Predicted, PredMeans, and Sample tables from SAS 6 no longer exist and have been replaced by output data sets; see descriptions of the MODEL statement options OUTP= and OUTPM= and the PRIOR statement option OUT= for more details. The ML and REML tables from SAS 6 have been replaced by the IterHistory table. The Tests, HLM, and HLPS tables from SAS 6 have been renamed Tests3, HLM3, and HLPS3, respectively.

Table 59.27 lists the variable names associated with the data sets created when you use the ODS OUTPUT option in conjunction with the preceding tables. In Table 59.27, n is used to denote a generic number that depends on the particular data set and model you select, and it can assume a different value each time it is used (even within the same table). The phrase model specific appears in rows of the affected tables to indicate that columns in these tables depend on the variables you specify in the model.

Caution: There is a danger of name collisions with the variables in the model specific tables in Table 59.27 and variables in your input data set. You should avoid using input variables with the same names as the variables in these tables.

Table 59.27: Variable Names for the ODS Tables Produced in PROC MIXED

Table Name

Variables

AsyCorr

Row, CovParm, CovP1–CovPn

AsyCov

Row, CovParm, CovP1–CovPn

Base

Type, Parm1–Parmn

Bound

Technique, Converge, Iterations, Evaluations, LogBound, CovP1–CovPn, TCovP1–TCovPn

CholG

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

CholR

Index, Row, Col1–Coln

CholV

Index, Row, Col1–Coln

ClassLevels

Class, Levels, Values

Coef

Model specific, LMatrix, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row1–Rown

Contrasts

Label, NumDF, DenDF, ChiSquare, FValue, ProbChiSq, ProbF

CorrB

Model specific, Effect, Row, Col1–Coln

CovB

Model specific, Effect, Row, Col1–Coln

CovParms

CovParm, Subject, Group, Estimate, StandardError, ZValue, ProbZ, Alpha, Lower, Upper

Diffs

Model specific, Effect, Margins, ByLevel, AT variables, Diff, StandardError, DF, tValue, Tails, Probt, Adjustment, Adjp, Alpha, Lower, Upper, AdjLow, AdjUpp

Dimensions

Descr, Value

Estimates

Label, Estimate, StandardError, DF, tValue, Tails, Probt, Alpha, Lower, Upper

FitStatistics

Descr, Value

G

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

GCorr

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

HLM1

Effect, NumDF, DenDF, FValue, ProbF

HLM2

Effect, NumDF, DenDF, FValue, ProbF

HLM3

Effect, NumDF, DenDF, FValue, ProbF

HLPS1

Effect, NumDF, DenDF, FValue, ProbF

HLPS2

Effect, NumDF, DenDF, FValue, ProbF

HLPS3

Effect, NumDF, DenDF, FValue, ProbF

Influence

Dependent on option modifiers, Effect, Tuple, Obs1–Obsk, Level, Iter, Index, Predicted, Residual, Leverage, PressRes, PRESS, Student, RMSE, RStudent, CookD, DFFITS, MDFFITS, CovRatio, CovTrace, CookDCP, MDFFITSCP, CovRatioCP, CovTraceCP, LD, RLD, Parm1–Parmp, CovP1–CovPq, Notes

InfoCrit

Neg2LogLike, Parms, AIC, AICC, HQIC, BIC, CAIC

InvCholG

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

InvCholR

Index, Row, Col1–Coln

InvCholV

Index, Row, Col1–Coln

InvCovB

Model specific, Effect, Row, Col1–Coln

InvG

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

InvR

Index, Row, Col1–Coln

InvV

Index, Row, Col1–Coln

IterHistory

CovP1–CovPn, Iteration, Evaluations, M2ResLogLike, M2LogLike, Criterion

LComponents

Effect, TestType, LIndex, Estimate, StdErr, DF, tValue, Probt

LRT

DF, ChiSquare, ProbChiSq

LSMeans

Model specific, Effect, Margins, ByLevel, AT variables, Estimate, StandardError, DF, tValue, Probt, Alpha, Lower, Upper, Cov1–Covn, Corr1–Corrn

MMEq

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

MMEqSol

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Row, Col1–Coln

ModelInfo

Descr, Value

Nobs

Label, N, NObsRead, NObsUsed, SumFreqsRead, SumFreqsUsed

ParmSearch

CovP1–CovPn, Var, ResLogLike, M2ResLogLike2, LogLike, M2LogLike, LogDetH

Posterior

Descr, Value

R

Index, Row, Col1–Coln

RCorr

Index, Row, Col1–Coln

Search

Parm, TCovP1–TCovPn, Posterior

Slices

Model specific, Effect, Margins, ByLevel, AT variables, NumDF, DenDF, FValue, ProbF

SolutionF

Model specific, Effect, Estimate, StandardError, DF, tValue, Probt, Alpha, Lower, Upper

SolutionR

Model specific, Effect, Subject, Sub1–Subn, Group, Group1–Groupn, Estimate, StdErrPred, DF, tValue, Probt, Alpha, Lower, Upper

Tests1

Effect, NumDF, DenDF, ChiSquare, FValue, ProbChiSq, ProbF

Tests2

Effect, NumDF, DenDF, ChiSquare, FValue, ProbChiSq, ProbF

Tests3

Effect, NumDF, DenDF, ChiSquare, FValue, ProbChiSq, ProbF

Type1

Source, DF, SS, MS, EMS, ErrorTerm, ErrorDF, FValue, ProbF

Type2

Source, DF, SS, MS, EMS, ErrorTerm, ErrorDF, FValue, ProbF

Type3

Source, DF, SS, MS, EMS, ErrorTerm, ErrorDF, FValue, ProbF

Trans

Prior, TCovP, CovP1–CovPn

V

Index, Row, Col1–Coln

VCorr

Index, Row, Col1–Coln


Some of the variables listed in Table 59.27 are created only when you specify certain options in the relevant PROC MIXED statements.