“Mplus ANALYSIS”的版本间的差异

2017年2月27日 (一) 14:11的版本 （查看源代码） Lichaoping（讨论 | 贡献） ←上一编辑		2017年2月27日 (一) 15:05的最新版本 （查看源代码） Lichaoping（讨论 | 贡献） （→‎ANALYSIS子命令）
（未显示同一用户的9个中间版本）
第1行：		第1行：
−	ANALYSIS命令语法	+	==ANALYSIS命令语法==
−	{| <tbody> style=" width: 100%;margin:0px; border: solid 1px #AAAAAA; border-spacing: 5px;"	+	*HTML版请访问：http://www.statmodel.com/HTML_UG/chapter16V8.htm
−	| width="168" | ANALYSIS:
−	| width="294" |
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" |
−	| width="120" |
−	|-
−	| width="168" | TYPE =
−	| width="294" | <strong>GEN</strong>ERAL;
−	| width="120" | GENERAL
−	|-
−	| width="168" |
−	| width="294" | <strong>    BAS</strong>IC;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>    RAND</strong>OM;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>    COM</strong>PLEX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>MIX</strong>TURE;
−	<strong>    BAS</strong>IC;
−	<strong>    RAND</strong>OM;
−	<strong>    COM</strong>PLEX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>TWO</strong>LEVEL;
−	<strong>    BAS</strong>IC;
−	<strong>    RAND</strong>OM;
−	<strong>    MIX</strong>TURE;
−	<strong>    COM</strong>PLEX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>THREE</strong>LEVEL;
−	<strong>    BAS</strong>IC;
−	<strong>    RAND</strong>OM;
−	<strong>    COM</strong>PLEX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CROSS</strong>CLASSIFIED;
−	<strong>    RAND</strong>OM;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | EFA # #;
−	<strong>    BAS</strong>IC;
−	<strong>    MIX</strong>TURE;
−	<strong>    COM</strong>PLEX;
−	<strong>    TWO</strong>LEVEL;
−	EFA # # UW* # # UB*;
−	EFA # # UW # # UB;
−	| width="120" |
−	|-
−	| width="168" | ESTIMATOR =
−	| width="294" | ML;
−	| width="120" | depends on
−	|-
−	| width="168" |
−	| width="294" | MLM;
−	| width="120" | analysis type
−	|-
−	| width="168" |
−	| width="294" | MLMV;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | MLR;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | MLF;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | MUML;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | WLS;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | WLSM;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | WLSMV;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | ULS;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | ULSMV;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | GLS;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | BAYES;
−	| width="120" |
−	|-
−	| width="168" | MODEL =
−	| width="294" | <strong>CONFIG</strong>URAL;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | METRIC;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | SCALAR;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>NOMEAN</strong>STRUCTURE;
−	| width="120" | means
−	|-
−	| width="168" |
−	| width="294" | <strong>NOCOV</strong>ARIANCES;
−	| width="120" | covariances
−	|-
−	| width="168" |
−	| width="294" | <strong>ALL</strong>FREE;
−	| width="120" | equal
−	|-
−	| width="168" | ALIGNMENT =
−	| width="294" | FIXED;
−	| width="120" | last class
−	|-
−	| width="168" |
−	| width="294" |
−	| width="120" | CONFIGURAL
−	|-
−	| width="168" |
−	| width="294" | FIXED (reference class <strong>CONFIG</strong>URAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | FIXED (reference class BSEM);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | FREE;
−	| width="120" | last class
−	|-
−	| width="168" |
−	| width="294" |
−	| width="120" | CONFIGURAL
−	|-
−	| width="168" |
−	| width="294" | FREE (reference class <strong>CONFIG</strong>URAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | FREE (reference class BSEM);
−	| width="120" |
−	|-
−	| width="168" | DISTRIBUTION =
−	| width="294" | <strong>NORM</strong>AL;
−	| width="120" | NORMAL
−	|-
−	| width="168" |
−	| width="294" | <strong>SKEW</strong>NORMAL;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>TDIST</strong>RIBUTION;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | SKEWT;
−	| width="120" |
−	|-
−	| width="168" | PARAMETERIZATION =
−	| width="294" | DELTA;
−	| width="120" | DELTA
−	|-
−	| width="168" |
−	| width="294" | THETA;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | LOGIT;
−	| width="120" | LOGIT
−	|-
−	| width="168" |
−	| width="294" | <strong>LOGLIN</strong>EAR;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>PROB</strong>ABILITY;
−	<strong>RESCOV</strong>ARIANCES;
−	| width="120" |
−	RESCOV
−	|-
−	| width="168" | LINK =
−	| width="294" | <strong>LOG</strong>IT;
−	| width="120" | LOGIT
−	|-
−	| width="168" |
−	| width="294" | <strong>PROB</strong>IT;
−	| width="120" |
−	|-
−	| width="168" | ROTATION =
−	| width="294" | <strong>GEO</strong>MIN;
−	| width="120" | GEOMIN (OBLIQUE value)
−	|-
−	| width="168" |
−	| width="294" | <strong>GEO</strong>MIN (<strong>OB</strong>LIQUE value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>GEO</strong>MIN (<strong>OR</strong>THOGONAL value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>QUART</strong>IMIN;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-V</strong>ARIMAX;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-V</strong>ARIMAX (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-V</strong>ARIMAX (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-Q</strong>UARTIMAX;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- Q</strong>UARTIMAX (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- Q</strong>UARTIMAX (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-E</strong>QUAMAX;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- E</strong>QUAMAX (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- E</strong>QUAMAX (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-P</strong>ARSIMAX;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- P</strong>ARSIMAX (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- P</strong>ARSIMAX (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF-F</strong>ACPARSIM;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- F</strong>ACPARSIM (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CF- F</strong>ACPARSIM (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CRAW</strong>FER;
−	| width="120" | OBLIQUE 1/p
−	|-
−	| width="168" |
−	| width="294" | <strong>CRAW</strong>FER (<strong>OB</strong>LIQUE value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>CRAW</strong>FER (<strong>OR</strong>THOGONAL value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>OBLIM</strong>IN;
−	| width="120" | OBLIQUE 0
−	|-
−	| width="168" |
−	| width="294" | <strong>OBLIM</strong>IN (<strong>OB</strong>LIQUE value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>OBLIM</strong>IN (<strong>OR</strong>THOGONAL value);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>VAR</strong>IMAX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>PRO</strong>MAX;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>TAR</strong>GET;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-GEO</strong>MIN;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-GEO</strong>MIN (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-GEO</strong>MIN (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-CF-Q</strong>UARTIMAX;
−	| width="120" | OBLIQUE
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-CF-Q</strong>UARTIMAX (<strong>OB</strong>LIQUE);
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>BI-CF-Q</strong>UARTIMAX (<strong>OR</strong>THOGONAL);
−	| width="120" |
−	|-
−	| width="168" | ROWSTANDARDIZATION =
−	| width="294" | <strong>CORR</strong>ELATION;
−	| width="120" | CORRELATION
−	|-
−	| width="168" |
−	| width="294" | <strong>KAIS</strong>ER;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | <strong>COVA</strong>RIANCE;
−	| width="120" |
−	|-
−	| width="168" | PARALLEL =
−	| width="294" | number;
−	| width="120" | 0
−	|-
−	| width="168" | REPSE =
−	| width="294" | <strong>BOOT</strong>STRAP;
−	<strong>JACK</strong>KNIFE;
−	<strong>JACK</strong>KNIFE<strong>1</strong>;
−	<strong>JACK</strong>KNIFE<strong>2</strong>;
−	BRR;
−	FAY (#);
−	| width="120" |
		+	==ANALYSIS子命令==
		+	ANALYSIS命令有77条子命令。
−	.3	+	* TYPE =
−	|-
−	| width="168" | BASEHAZARD =
−	| width="294" | ON;
−	OFF;
−	ON (<strong>EQ</strong>UAL);
−	ON (<strong>UNEQ</strong>UAL);
−	OFF (<strong>EQ</strong>UAL);
−	OFF (<strong>UNEQ</strong>UAL);
−	| width="120" | depends on
−	analysis type
−	EQUAL
−	EQUAL	+	* ESTIMATOR =
−	|-
−	| width="168" | CHOLESKY =
−	| width="294" | ON;
−	OFF;
−	| width="120" | depends on
−	analysis type
−	|-
−	| width="168" | ALGORITHM =
−	| width="294" | EM;
−	| width="120" | depends on
−	|-
−	| width="168" |
−	| width="294" | EMA;
−	| width="120" | analysis type
−	|-
−	| width="168" |
−	| width="294" | FS;
−	ODLL;
−	<strong>INT</strong>EGRATION;
−	| width="120" |
−	|-
−	| width="168" | INTEGRATION =
−	| width="294" | number of integration points;
−	<strong>STAND</strong>ARD (number of integration points) ;
−	<strong>GAUSS</strong>HERMITE (number of integration points) ;	+	* MODEL =
−	<strong>MONTE</strong>CARLO (number of integration points);
−	| width="120" | STANDARD
−	depends on
−	analysis type
−	15
−	depends on	+	* ALIGNMENT =
−	analysis type	+
−	|-	+	* DISTRIBUTION =
−	| width="168" | MCSEED =	+
−	| width="294" | random seed for Monte Carlo integration;	+	* PARAMETERIZATION =
−	| width="120" | 0	+
−	|-	+	* LINK =
−	| width="168" | ADAPTIVE =	+
−	| width="294" | ON;	+	* ROTATION =
−	OFF;	+
−	| width="120" | ON	+	* ROWSTANDARDIZATION =
−	|-	+
−	| width="168" | INFORMATION =	+	* PARALLEL =
−	| width="294" | <strong>OBS</strong>ERVED;	+
−	| width="120" | depends on	+	* REPSE =
−	|-	+
−	| width="168" |	+	* BASEHAZARD =
−	| width="294" | <strong>EXP</strong>ECTED;	+
−	| width="120" | analysis type	+	* CHOLESKY =
−	|-	+
−	| width="168" |	+	* ALGORITHM =
−	| width="294" | <strong>COMB</strong>INATION;	+
−	| width="120" |	+	* INTEGRATION =
−	|-	+
−	| width="168" | BOOTSTRAP =	+	* MCSEED =
−	| width="294" | number of bootstrap draws;	+
−	number of bootstrap draws (<strong>STAND</strong>ARD);	+	* ADAPTIVE =
−	number of bootstrap draws (<strong>RES</strong>IDUAL):	+
−	| width="120" | STANDARD	+	* INFORMATION =
−	|-	+
−	| width="168" | LRTBOOTSTRAP =	+	* BOOTSTRAP =
−	| width="294" | number of bootstrap draws for TECH14;	+
−	| width="120" | depends on	+	* LRTBOOTSTRAP =
−	analysis type	+
−	|-	+	* STARTS =
−	| width="168" | STARTS =	+
−	| width="294" | number of initial stage starts and number of final stage optimizations;	+	* STITERATIONS =
−	| width="120" | depends on	+
−	analysis type	+	* STCONVERGENCE =
−	|-	+
−	| width="168" | STITERATIONS =	+	* STSCALE =
−	| width="294" | number of initial stage iterations;	+
−	| width="120" | 10	+	* STSEED =
−	|-	+
−	| width="168" | STCONVERGENCE =	+	* OPTSEED =
−	| width="294" | initial stage convergence criterion;	+
−	| width="120" | 1	+	* K-1STARTS =
−	|-	+
−	| width="168" | STSCALE =	+	* LRTSTARTS =
−	| width="294" | random start scale;	+
−	| width="120" | 5	+	* RSTARTS =
−	|-	+
−	| width="168" | STSEED =	+	* ASTARTS =
−	| width="294" | random seed for generating random starts;	+
−	| width="120" | 0	+	* H1STARTS =
−	|-	+
−	| width="168" | OPTSEED =	+	* DIFFTEST =
−	| width="294" | random seed for analysis;	+
−	| width="120" |	+	* MULTIPLIER =
−	|-	+
−	| width="168" | K-1STARTS =	+	* COVERAGE =
−	| width="294" | number of initial stage starts and number of final stage optimizations for the k-1 class model for TECH14;	+
−	| width="120" | 20 4	+	* ADDFREQUENCY =
−	|-	+
−	| width="168" | LRTSTARTS =	+	* ITERATIONS =
−	| width="294" | number of initial stage starts and number of final stage optimizations for TECH14;	+
−	| width="120" | 0 0 40 8	+	* SDITERATIONS =
−	|-	+
−	| width="168" | RSTARTS =	+	* H1ITERATIONS =
−	| width="294" | number of random starts for the rotation algorithm and number of factor solutions printed for exploratory factor analysis;	+
−	| width="120" | depends on	+	* MITERATIONS =
−	analysis type	+
−	|-	+	* MCITERATIONS =
−	| width="168" | ASTARTS =	+
−	| width="294" | number of random starts for the alignment	+	* MUITERATIONS =
−	optimization;	+
−	| width="120" | 30	+	* RITERATIONS =
−	|-	+
−	| width="168" | H1STARTS =	+	* AITERATIONS =
−	| width="294" | Number of initial stage starts and number of final stage optimizations for the H1 model;	+
−	| width="120" | 0 0	+	* CONVERGENCE =
−	|-	+
−	| width="168" | DIFFTEST =	+	* H1CONVERGENCE =
−	| width="294" | file name;	+
−	| width="120" |	+	* LOGCRITERION =
−	|-	+
−	| width="168" | MULTIPLIER =	+	* RLOGCRITERION =
−	| width="294" | file name;	+
−	| width="120" |	+	* MCONVERGENCE =
−	|-	+
−	| width="168" | COVERAGE =	+	* MCCONVERGENCE =
−	| width="294" | minimum covariance coverage with missing data;	+
−	| width="120" | .10	+	* MUCONVERGENCE =
−	|-	+
−	| width="168" | ADDFREQUENCY =	+	* RCONVERGENCE =
−	| width="294" | value divided by sample size to add to cells with zero frequency;	+
−	| width="120" | .5	+	* ACONVERGENCE =
−	|-	+
−	| width="168" | ITERATIONS =	+	* MIXC =
−	| width="294" | maximum number of iterations for the Quasi-Newton algorithm for continuous outcomes;	+
−	| width="120" | 1000	+	* MIXU =
−	|-	+
−	| width="168" | SDITERATIONS =	+	* LOGHIGH =
−	| width="294" | maximum number of steepest descent iterations for the Quasi-Newton algorithm for continuous outcomes;	+
−	| width="120" | 20	+	* LOGLOW =
−	|-	+
−	| width="168" | H1ITERATIONS =	+	* UCELLSIZE =
−	| width="294" | maximum number of iterations for unrestricted model with missing data;	+
−	| width="120" | 2000	+	* VARIANCE =
−	|-	+
−	| width="168" | MITERATIONS =	+	* SIMPLICITY =
−	| width="294" | number of iterations for the EM algorithm;	+
−	| width="120" | 500	+	* TOLERANCE =
−	|-	+
−	| width="168" | MCITERATIONS =	+	* METRIC=
−	| width="294" | number of iterations for the M step of the EM algorithm for categorical latent variables;	+
−	| width="120" | 1	+	* MATRIX =
−	|-	+
−	| width="168" | MUITERATIONS =	+	* POINT =
−	| width="294" | number of iterations for the M step of the EM algorithm for censored, categorical, and count outcomes;	+
−	| width="120" | 1	+	* CHAINS =
−	|-	+
−	| width="168" | RITERATIONS =	+	* BSEED =
−	| width="294" | maximum number of iterations in the rotation algorithm for exploratory factor analysis;	+
−	| width="120" | 10000	+	* STVALUES =
−	|-	+
−	| width="168" | AITERATIONS =	+	* MEDIATOR =
−	| width="294" | maximum number of iterations in the	+
−	| width="120" | 5000	+	* ALGORITHM =
−	|-	+
−	| width="168" |	+	* BCONVERGENCE =
−	| width="294" | alignment optimization;	+
−	| width="120" |	+	* BITERATIONS =
−	|-	+
−	| width="168" | CONVERGENCE =	+	* FBITERATIONS =
−	| width="294" | convergence criterion for the Quasi-Newton algorithm for continuous outcomes;	+
−	| width="120" | depends on	+	* THIN =
−	analysis type	+
−	|-	+	* MDITERATIONS =
−	| width="168" | H1CONVERGENCE =	+
−	| width="294" | convergence criterion for unrestricted model with missing data;	+	* KOLMOGOROV =
−	| width="120" | .0001	+
−	|-	+	* PRIOR =
−	| width="168" | LOGCRITERION =	+
−	| width="294" | likelihood convergence criterion for the EM algorithm;	+	* INTERACTIVE =
−	| width="120" | depends on	+
−	analysis type	+	* PROCESSORS =
−	|-
−	| width="168" | RLOGCRITERION =
−	| width="294" | relative likelihood convergence criterion for the EM algorithm;
−	| width="120" | depends on
−	analysis type
−	|-
−	| width="168" | MCONVERGENCE =
−	| width="294" | convergence criterion for the EM algorithm;
−	| width="120" | depends on
−	analysis type
−	|-
−	| width="168" | MCCONVERGENCE =
−	| width="294" | convergence criterion for the M step of the EM algorithm for categorical latent variables;
−	| width="120" | .000001
−	|-
−	| width="168" | MUCONVERGENCE =
−	| width="294" | convergence criterion for the M step of the EM algorithm for censored, categorical, and count outcomes;
−	| width="120" | .000001
−	|-
−	| width="168" | RCONVERGENCE =
−	| width="294" | convergence criterion for the rotation algorithm for exploratory factor analysis;
−	| width="120" | .00001
−	|-
−	| width="168" | ACONVERGENCE =
−	| width="294" | convergence criterion for the derivatives of
−	the alignment optimization;.
−	| width="120" | .001
−	|-
−	| width="168" | MIXC =
−	| width="294" | <strong>ITER</strong>ATIONS;
−	| width="120" | ITERATIONS
−	|-
−	| width="168" |
−	| width="294" | <strong>CONV</strong>ERGENCE;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | M step iteration termination based on number of iterations or convergence for categorical latent variables;
−	| width="120" |
−	|-
−	| width="168" | MIXU =
−	| width="294" | <strong>ITER</strong>ATIONS;
−	| width="120" | ITERATIONS
−	|-
−	| width="168" |
−	| width="294" | <strong>CONV</strong>ERGENCE;
−	| width="120" |
−	|-
−	| width="168" |
−	| width="294" | M step iteration termination based on number of iterations or convergence for censored, categorical, and count outcomes;
−	| width="120" |
−	|-
−	| width="168" | LOGHIGH =
−	| width="294" | max value for logit thresholds;
−	| width="120" | +15
−	|-
−	| width="168" | LOGLOW =
−	| width="294" | min value for logit thresholds;
−	| width="120" | - 15
−	|-
−	| width="168" | UCELLSIZE =
−	| width="294" | minimum expected cell size;
−	| width="120" | .01
−	|-
−	| width="168" | VARIANCE =
−	| width="294" | minimum variance value;
−	| width="120" | .0001
−	|-
−	| width="168" | SIMPLICITY =
−	| width="294" | SQRT;
−	| width="120" | SQRT
−	|-
−	| width="168" |
−	| width="294" | <strong>FOUR</strong>THRT;
−	| width="120" |
−	|-
−	| width="168" | TOLERANCE =
−	| width="294" | simplicity tolerance value;
−	| width="120" | .0001
−	|-
−	| width="168" | METRIC=
−	| width="294" | <strong>REFG</strong>ROUP;
−	| width="120" | REFGROUP
−	|-
−	| width="168" |
−	| width="294" | <strong>PROD</strong>UCT;
−	| width="120" |
−	|-
−	| width="168" | MATRIX =
−	| width="294" | <strong>COVA</strong>RIANCE;
−	| width="120" | COVARIANCE
−	|-
−	| width="168" |
−	| width="294" | <strong>CORR</strong>ELATION;
−	| width="120" |
−	|-
−	| width="168" | POINT =
−	| width="294" | <strong>MED</strong>IAN;
−	MEAN;
−	MODE;
−	| width="120" | MEDIAN
−	|-
−	| width="168" | CHAINS =
−	| width="294" | number of MCMC chains;
−	| width="120" | 2
−	|-
−	| width="168" | BSEED =
−	| width="294" | seed for MCMC random number generation;
−	| width="120" | 0
−	|-
−	| width="168" | STVALUES =
−	| width="294" | <strong>UNPER</strong>TURBED;
−	<strong>PERT</strong>URBED;
−	ML;
−	| width="120" | UNPERTURBED
−	|-
−	| width="168" | MEDIATOR =
−	| width="294" | <strong>LAT</strong>ENT;
−	<strong>OBS</strong>ERVED;
−	| width="120" | depends on
−	analysis type
−	|-
−	| width="168" | ALGORITHM =
−	| width="294" | GIBBS;
−	GIBBS (PX1);
−	GIBBS (PX2);
−	GIBBS (PX3);
−	GIBBS (RW);
−	MH;
−	| width="120" | GIBBS (PX1)
−	|-
−	| width="168" | BCONVERGENCE =
−	| width="294" | MCMC convergence criterion using Gelman-Rubin PSR;
−	| width="120" | .05
−	|-
−	| width="168" | BITERATIONS =
−	| width="294" | maximum and minimum number of iterations for each MCMC chain when Gelman-Rubin PSR is used;
−	| width="120" | 50000 0
−	|-
−	| width="168" | FBITERATIONS =
−	| width="294" | fixed number of iterations for each MCMC chain when Gelman-Rubin PSR is not used;
−	| width="120" |
−	|-
−	| width="168" | THIN =
−	| width="294" | k where every k-th MCMC iteration is saved;
−	| width="120" | 1
−	|-
−	| width="168" | MDITERATIONS =
−	| width="294" | maximum number of iterations used to compute the Bayes multivariate mode;
−	| width="120" | 10000
−	|-
−	| width="168" | KOLMOGOROV =
−	| width="294" | number of draws from the MCMC chains;
−	| width="120" | 100
−	|-
−	| width="168" | PRIOR =
−	| width="294" | number of draws from the prior distribution;
−	| width="120" | 1000
−	|-
−	| width="168" | INTERACTIVE =
−	| width="294" | file name;
−	| width="120" |
−	|-
−	| width="168" | PROCESSORS =
−	| width="294" | # of processors # of threads;
−	| width="120" | 1 1</tbody>
−	|}

---

2017年2月27日 (一) 15:05的最新版本

ANALYSIS命令语法

• HTML版请访问：http://www.statmodel.com/HTML_UG/chapter16V8.htm

ANALYSIS子命令

ANALYSIS命令有77条子命令。

• TYPE =

• ESTIMATOR =

• MODEL =

• ALIGNMENT =

• DISTRIBUTION =

• PARAMETERIZATION =

• LINK =

• ROTATION =

• ROWSTANDARDIZATION =

• PARALLEL =

• REPSE =

• BASEHAZARD =

• CHOLESKY =

• ALGORITHM =

• INTEGRATION =

• MCSEED =

• ADAPTIVE =

• INFORMATION =

• BOOTSTRAP =

• LRTBOOTSTRAP =

• STARTS =

• STITERATIONS =

• STCONVERGENCE =

• STSCALE =

• STSEED =

• OPTSEED =

• K-1STARTS =

• LRTSTARTS =

• RSTARTS =

• ASTARTS =

• H1STARTS =

• DIFFTEST =

• MULTIPLIER =

• COVERAGE =

• ADDFREQUENCY =

• ITERATIONS =

• SDITERATIONS =

• H1ITERATIONS =

• MITERATIONS =

• MCITERATIONS =

• MUITERATIONS =

• RITERATIONS =

• AITERATIONS =

• CONVERGENCE =

• H1CONVERGENCE =

• LOGCRITERION =

• RLOGCRITERION =

• MCONVERGENCE =

• MCCONVERGENCE =

• MUCONVERGENCE =

• RCONVERGENCE =

• ACONVERGENCE =

• MIXC =

• MIXU =

• LOGHIGH =

• LOGLOW =

• UCELLSIZE =

• VARIANCE =

• SIMPLICITY =

• TOLERANCE =

• METRIC=

• MATRIX =

• POINT =

• CHAINS =

• BSEED =

• STVALUES =

• MEDIATOR =

• ALGORITHM =

• BCONVERGENCE =

• BITERATIONS =

• FBITERATIONS =

• THIN =

• MDITERATIONS =

• KOLMOGOROV =

• PRIOR =

• INTERACTIVE =

• PROCESSORS =