Mplus VERSION 7.4
MUTHEN & MUTHEN
06/02/2016 8:27 PM
INPUT INSTRUCTIONS
Title:
Clinical Trials data
M (intent) original 4-category mediator version
data:
file = 4cat m.dat;
variable:
names = intent tx ciguse w;
usev = tx ciguse intent;
categorical = ciguse intent;
freqweight = w;
Analysis:
estimator = bayes;
processors = 2;
biteration = (10000);
model:
ciguse on tx intent;
intent on tx;
model indirect: ciguse ind intent tx;
output:
tech1 tech8 sampstat patterns cinterval;
plot:
type = plot3;
*** WARNING in OUTPUT command
SAMPSTAT option is not available for ESTIMATOR=BAYES. Use TYPE=BASIC.
Request for SAMPSTAT is ignored.
*** WARNING in OUTPUT command
PATTERNS option is available only when there are missing data.
Request for PATTERNS is ignored.
2 WARNING(S) FOUND IN THE INPUT INSTRUCTIONS
Clinical Trials data
M (intent) original 4-category mediator version
SUMMARY OF ANALYSIS
Number of groups 1
Number of observations 864
Number of dependent variables 2
Number of independent variables 1
Number of continuous latent variables 0
Observed dependent variables
Binary and ordered categorical (ordinal)
CIGUSE INTENT
Observed independent variables
TX
Estimator BAYES
Specifications for Bayesian Estimation
Point estimate MEDIAN
Number of Markov chain Monte Carlo (MCMC) chains 2
Random seed for the first chain 0
Starting value information UNPERTURBED
Treatment of categorical mediator LATENT
Algorithm used for Markov chain Monte Carlo GIBBS(PX1)
Convergence criterion 0.500D-01
Maximum number of iterations 50000
K-th iteration used for thinning 1
Input data file(s)
4cat m.dat
Input data format FREE
UNIVARIATE PROPORTIONS AND COUNTS FOR CATEGORICAL VARIABLES
CIGUSE
Category 1 0.819 708.000
Category 2 0.181 156.000
INTENT
Category 1 0.745 644.000
Category 2 0.119 103.000
Category 3 0.069 60.000
Category 4 0.066 57.000
THE MODEL ESTIMATION TERMINATED NORMALLY
USE THE FBITERATIONS OPTION TO INCREASE THE NUMBER OF ITERATIONS BY A FACTOR
OF AT LEAST TWO TO CHECK CONVERGENCE AND THAT THE PSR VALUE DOES NOT INCREASE.
MODEL FIT INFORMATION
Number of Free Parameters 7
Bayesian Posterior Predictive Checking using Chi-Square
95% Confidence Interval for the Difference Between
the Observed and the Replicated Chi-Square Values
-10.936 11.911
Posterior Predictive P-Value 0.497
MODEL RESULTS
Posterior One-Tailed 95% C.I.
Estimate S.D. P-Value Lower 2.5% Upper 2.5% Significance
CIGUSE ON
TX -0.175 0.118 0.072 -0.405 0.057
INTENT 0.810 0.090 0.000 0.643 0.999 *
INTENT ON
TX -0.235 0.087 0.003 -0.409 -0.067 *
Thresholds
CIGUSE$1 0.977 0.102 0.000 0.786 1.189 *
INTENT$1 0.525 0.065 0.000 0.391 0.651 *
INTENT$2 0.976 0.070 0.000 0.830 1.112 *
INTENT$3 1.396 0.080 0.000 1.245 1.563 *
TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS FOR LATENT RESPONSE VARIABLES
Posterior One-Tailed 95% C.I.
Estimate S.D. P-Value Lower 2.5% Upper 2.5% Significance
Effects from TX to CIGUSE
Indirect -0.189 0.074 0.003 -0.344 -0.053 *
Direct effect -0.175 0.118 0.072 -0.405 0.057
TOTAL, INDIRECT, AND DIRECT EFFECTS BASED ON COUNTERFACTUALS (CAUSALLY-DEFINED EFFECTS)
Posterior One-Tailed 95% C.I.
Estimate S.D. P-Value Lower 2.5% Upper 2.5% Significance
Effects from TX to CIGUSE
Tot natural IE -0.028 0.011 0.003 -0.053 -0.008 *
Pure natural DE -0.046 0.025 0.033 -0.097 0.003
Total effect -0.075 0.027 0.002 -0.129 -0.023 *
Odds ratios for binary Y
Tot natural IE 0.810 0.066 0.000 0.683 0.943 *
Pure natural DE 0.748 0.121 0.000 0.547 1.017 *
Total effect 0.604 0.110 0.000 0.425 0.858 *
Other effects
Pure natural IE -0.031 0.012 0.003 -0.056 -0.009 *
Tot natural DE -0.044 0.024 0.028 -0.092 0.001
Total effect -0.075 0.027 0.002 -0.129 -0.023 *
Odds ratios for other effects for binary Y
Pure natural IE 0.829 0.059 0.000 0.717 0.947 *
Tot natural DE 0.730 0.124 0.000 0.526 1.008 *
Total effect 0.604 0.110 0.000 0.425 0.858 *
CREDIBILITY INTERVALS OF MODEL RESULTS
Lower .5% Lower 2.5% Lower 5% Estimate Upper 5% Upper 2.5% Upper .5%
CIGUSE ON
TX -0.478 -0.405 -0.368 -0.175 0.020 0.057 0.141
INTENT 0.597 0.643 0.668 0.810 0.968 0.999 1.054
INTENT ON
TX -0.463 -0.409 -0.382 -0.235 -0.094 -0.067 -0.015
Thresholds
CIGUSE$1 0.723 0.786 0.815 0.977 1.151 1.189 1.262
INTENT$1 0.350 0.391 0.414 0.525 0.633 0.651 0.689
INTENT$2 0.783 0.830 0.856 0.976 1.087 1.112 1.153
INTENT$3 1.183 1.245 1.271 1.396 1.536 1.563 1.614
CREDIBILITY INTERVALS OF TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS FOR LATENT RESPONSE VARIABLES
Lower .5% Lower 2.5% Lower 5% Estimate Upper 5% Upper 2.5% Upper .5%
Effects from TX to CIGUSE
Indirect -0.402 -0.344 -0.317 -0.189 -0.075 -0.053 -0.013
Direct effect -0.478 -0.405 -0.368 -0.175 0.020 0.057 0.141
CREDIBILITY INTERVALS OF TOTAL, INDIRECT, AND DIRECT EFFECTS BASED ON COUNTERFACTUALS (CAUSALLY-DEFINED EFFECTS)
Lower .5% Lower 2.5% Lower 5% Estimate Upper 5% Upper 2.5% Upper .5%
Effects from TX to CIGUSE
Tot natural IE -0.061 -0.053 -0.049 -0.028 -0.011 -0.008 -0.002
Pure natural DE -0.112 -0.097 -0.089 -0.046 -0.005 0.003 0.019
Total effect -0.148 -0.129 -0.120 -0.075 -0.031 -0.023 -0.007
Odds ratios for binary Y
Tot natural IE 0.641 0.683 0.702 0.810 0.921 0.943 0.986
Pure natural DE 0.496 0.547 0.576 0.748 0.971 1.017 1.129
Total effect 0.384 0.425 0.453 0.604 0.811 0.858 0.953
Other effects
Pure natural IE -0.064 -0.056 -0.051 -0.031 -0.012 -0.009 -0.002
Tot natural DE -0.105 -0.092 -0.084 -0.044 -0.005 0.001 0.016
Total effect -0.148 -0.129 -0.120 -0.075 -0.031 -0.023 -0.007
Odds ratios for other effects for binary Y
Pure natural IE 0.682 0.717 0.735 0.829 0.928 0.947 0.987
Tot natural DE 0.476 0.526 0.556 0.730 0.961 1.008 1.122
Total effect 0.384 0.425 0.453 0.604 0.811 0.858 0.953
TECHNICAL 1 OUTPUT
PARAMETER SPECIFICATION
TAU
CIGUSE$1 INTENT$1 INTENT$2 INTENT$3
________ ________ ________ ________
1 4 5 6 7
NU
CIGUSE INTENT TX
________ ________ ________
1 0 0 0
LAMBDA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0 0 0
INTENT 0 0 0
TX 0 0 0
THETA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0
INTENT 0 0
TX 0 0 0
ALPHA
CIGUSE INTENT TX
________ ________ ________
1 0 0 0
BETA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0 1 2
INTENT 0 0 3
TX 0 0 0
PSI
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0
INTENT 0 0
TX 0 0 0
STARTING VALUES
TAU
CIGUSE$1 INTENT$1 INTENT$2 INTENT$3
________ ________ ________ ________
1 0.840 0.597 1.030 1.472
NU
CIGUSE INTENT TX
________ ________ ________
1 0.000 0.000 0.000
LAMBDA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 1.000 0.000 0.000
INTENT 0.000 1.000 0.000
TX 0.000 0.000 1.000
THETA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0.000
INTENT 0.000 0.000
TX 0.000 0.000 0.000
ALPHA
CIGUSE INTENT TX
________ ________ ________
1 0.000 0.000 0.000
BETA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0.000 0.000 0.000
INTENT 0.000 0.000 0.000
TX 0.000 0.000 0.000
PSI
CIGUSE INTENT TX
________ ________ ________
CIGUSE 1.000
INTENT 0.000 1.000
TX 0.000 0.000 0.123
PRIORS FOR ALL PARAMETERS PRIOR MEAN PRIOR VARIANCE PRIOR STD. DEV.
Parameter 1~N(0.000,5.000) 0.0000 5.0000 2.2361
Parameter 2~N(0.000,5.000) 0.0000 5.0000 2.2361
Parameter 3~N(0.000,5.000) 0.0000 5.0000 2.2361
Parameter 4~N(0.000,5.000) 0.0000 5.0000 2.2361
Parameter 5~N(0.000,infinity) 0.0000 infinity infinity
Parameter 6~N(0.000,infinity) 0.0000 infinity infinity
Parameter 7~N(0.000,infinity) 0.0000 infinity infinity
TECHNICAL 8 OUTPUT
Kolmogorov-Smirnov comparing posterior distributions across chains 1 and 2 using 100 draws.
Parameter KS Statistic P-value
Parameter 3 0.1600 0.1400
Parameter 2 0.1000 0.6766
Parameter 4 0.0900 0.7942
Parameter 7 0.0500 0.9995
Parameter 1 0.0400 1.0000
Parameter 5 0.0400 1.0000
Parameter 6 0.0400 1.0000
Simulated prior distributions
Parameter Prior Mean Prior Variance Prior Std. Dev.
Parameter 1 0.0149 5.3157 2.3056
Parameter 2 -0.0292 5.0725 2.2522
Parameter 3 0.0094 4.6021 2.1453
Parameter 4 -0.0292 4.8015 2.1912
Parameter 5 Improper Prior
Parameter 6 Improper Prior
Parameter 7 Improper Prior
TECHNICAL 8 OUTPUT FOR BAYES ESTIMATION
CHAIN BSEED
1 0
2 285380
POTENTIAL PARAMETER WITH
ITERATION SCALE REDUCTION HIGHEST PSR
100 1.045 5
200 1.535 5
300 1.120 4
400 1.164 5
500 1.252 6
600 1.424 6
700 1.148 6
800 1.221 6
900 1.062 7
1000 1.053 6
1100 1.059 6
1200 1.050 6
1300 1.019 6
1400 1.025 6
1500 1.025 6
1600 1.063 6
1700 1.090 6
1800 1.138 6
1900 1.139 6
2000 1.126 6
2100 1.125 6
2200 1.135 7
2300 1.110 7
2400 1.149 7
2500 1.178 7
2600 1.296 7
2700 1.311 7
2800 1.182 7
2900 1.069 7
3000 1.038 7
3100 1.019 7
3200 1.016 1
3300 1.012 1
3400 1.009 1
3500 1.009 1
3600 1.005 1
3700 1.011 5
3800 1.013 6
3900 1.015 6
4000 1.008 6
4100 1.007 6
4200 1.012 5
4300 1.018 7
4400 1.013 7
4500 1.006 7
4600 1.006 7
4700 1.009 7
4800 1.013 7
4900 1.014 7
5000 1.013 7
5100 1.012 4
5200 1.013 6
5300 1.015 6
5400 1.016 7
5500 1.011 7
5600 1.009 6
5700 1.009 6
5800 1.012 6
5900 1.006 6
6000 1.003 6
6100 1.003 4
6200 1.002 4
6300 1.002 2
6400 1.002 2
6500 1.003 2
6600 1.002 2
6700 1.002 1
6800 1.003 2
6900 1.002 2
7000 1.002 2
7100 1.003 2
7200 1.003 2
7300 1.003 2
7400 1.003 2
7500 1.003 2
7600 1.003 2
7700 1.003 2
7800 1.003 2
7900 1.003 2
8000 1.003 2
8100 1.002 2
8200 1.002 2
8300 1.002 2
8400 1.002 2
8500 1.002 2
8600 1.002 6
8700 1.003 6
8800 1.003 6
8900 1.003 6
9000 1.005 6
9100 1.007 6
9200 1.004 6
9300 1.001 2
9400 1.002 6
9500 1.004 6
9600 1.004 6
9700 1.001 6
9800 1.001 6
9900 1.001 1
10000 1.001 1
PLOT INFORMATION
The following plots are available:
Histograms (sample values)
Scatterplots (sample values)
Sample proportions and estimated probabilities
Bayesian posterior parameter distributions
Bayesian posterior parameter trace plots
Bayesian autocorrelation plots
Bayesian prior parameter distributions
Bayesian posterior predictive checking scatterplots
Bayesian posterior predictive checking distribution plots
DIAGRAM INFORMATION
Use View Diagram under the Diagram menu in the Mplus Editor to view the diagram.
If running Mplus from the Mplus Diagrammer, the diagram opens automatically.
Diagram output
c:\users\gryphon\desktop\chapter8\ex8.29_a4.dgm
Beginning Time: 20:27:45
Ending Time: 20:27:50
Elapsed Time: 00:00:05
MUTHEN & MUTHEN
3463 Stoner Ave.
Los Angeles, CA 90066
Tel: (310) 391-9971
Fax: (310) 391-8971
Web: www.StatModel.com
Support: Support@StatModel.com
Copyright (c) 1998-2015 Muthen & Muthen
Back to examples