Mplus VERSION 7.4
MUTHEN & MUTHEN
06/02/2016 5:03 PM
INPUT INSTRUCTIONS
Title:
Clinical Trials data
from MacKinnon et al. (2007)
data:
file = smoking.txt;
variable:
names = intent tx ciguse;
usev = tx ciguse intent;
categorical = ciguse;
analysis:
estimator = ml;
link = logit;
bootstrap = 10000;
model:
ciguse on intent (beta1)
tx (beta2);
intent on tx (gamma);
model indirect:
ciguse IND intent tx;
model constraint:
new(indirect direct);
indirect = exp(beta1*gamma);
direct = exp(beta2);
output:
tech1 tech8 sampstat cinterval(bootstrap);
plot:
type = plot3;
INPUT READING TERMINATED NORMALLY
Clinical Trials data
from MacKinnon et al. (2007)
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
Continuous
INTENT
Binary and ordered categorical (ordinal)
CIGUSE
Observed independent variables
TX
Estimator ML
Information matrix OBSERVED
Optimization Specifications for the Quasi-Newton Algorithm for
Continuous Outcomes
Maximum number of iterations 100
Convergence criterion 0.100D-05
Optimization Specifications for the EM Algorithm
Maximum number of iterations 500
Convergence criteria
Loglikelihood change 0.100D-02
Relative loglikelihood change 0.100D-05
Derivative 0.100D-02
Optimization Specifications for the M step of the EM Algorithm for
Categorical Latent variables
Number of M step iterations 1
M step convergence criterion 0.100D-02
Basis for M step termination ITERATION
Optimization Specifications for the M step of the EM Algorithm for
Censored, Binary or Ordered Categorical (Ordinal), Unordered
Categorical (Nominal) and Count Outcomes
Number of M step iterations 1
M step convergence criterion 0.100D-02
Basis for M step termination ITERATION
Maximum value for logit thresholds 15
Minimum value for logit thresholds -15
Minimum expected cell size for chi-square 0.100D-01
Number of bootstrap draws
Requested 10000
Completed 10000
Optimization algorithm EMA
Integration Specifications
Type STANDARD
Number of integration points 15
Dimensions of numerical integration 0
Adaptive quadrature ON
Link LOGIT
Cholesky OFF
Input data file(s)
smoking.txt
Input data format FREE
UNIVARIATE PROPORTIONS AND COUNTS FOR CATEGORICAL VARIABLES
CIGUSE
Category 1 0.819 708.000
Category 2 0.181 156.000
SAMPLE STATISTICS
SAMPLE STATISTICS
Means
INTENT TX
________ ________
1 1.456 0.571
Covariances
INTENT TX
________ ________
INTENT 0.783
TX -0.040 0.245
Correlations
INTENT TX
________ ________
INTENT 1.000
TX -0.092 1.000
UNIVARIATE SAMPLE STATISTICS
UNIVARIATE HIGHER-ORDER MOMENT DESCRIPTIVE STATISTICS
Variable/ Mean/ Skewness/ Minimum/ % with Percentiles
Sample Size Variance Kurtosis Maximum Min/Max 20%/60% 40%/80% Median
INTENT 1.456 1.863 1.000 74.54% 1.000 1.000 1.000
864.000 0.783 2.223 4.000 6.60% 1.000 2.000
TX 0.571 -0.285 0.000 42.94% 0.000 0.000 1.000
864.000 0.245 -1.919 1.000 57.06% 1.000 1.000
THE MODEL ESTIMATION TERMINATED NORMALLY
MODEL FIT INFORMATION
Number of Free Parameters 6
Loglikelihood
H0 Value -1452.174
Information Criteria
Akaike (AIC) 2916.348
Bayesian (BIC) 2944.918
Sample-Size Adjusted BIC 2925.863
(n* = (n + 2) / 24)
MODEL RESULTS
Two-Tailed
Estimate S.E. Est./S.E. P-Value
CIGUSE ON
INTENT 1.038 0.096 10.823 0.000
TX -0.377 0.203 -1.861 0.063
INTENT ON
TX -0.164 0.061 -2.681 0.007
Intercepts
INTENT 1.550 0.049 31.414 0.000
Thresholds
CIGUSE$1 3.056 0.225 13.573 0.000
Residual Variances
INTENT 0.776 0.055 14.190 0.000
New/Additional Parameters
INDIRECT 0.843 0.055 15.234 0.000
DIRECT 0.686 0.143 4.794 0.000
LOGISTIC REGRESSION ODDS RATIO RESULTS
CIGUSE ON
INTENT 2.823
TX 0.686
TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS FOR LATENT RESPONSE VARIABLES
Two-Tailed
Estimate S.E. Est./S.E. P-Value
Effects from TX to CIGUSE
Indirect -0.171 0.066 -2.592 0.010
Direct effect -0.377 0.203 -1.861 0.063
TOTAL, INDIRECT, AND DIRECT EFFECTS BASED ON COUNTERFACTUALS (CAUSALLY-DEFINED EFFECTS)
Two-Tailed
Estimate S.E. Est./S.E. P-Value
Effects from TX to CIGUSE
Tot natural IE -0.021 0.008 -2.526 0.012
Pure natural DE -0.053 0.028 -1.859 0.063
Total effect -0.073 0.029 -2.540 0.011
Odds ratios for binary Y
Tot natural IE 0.858 0.050 17.222 0.000
Pure natural DE 0.716 0.131 5.452 0.000
Total effect 0.614 0.120 5.138 0.000
Other effects
Pure natural IE -0.025 0.010 -2.556 0.011
Tot natural DE -0.048 0.026 -1.859 0.063
Total effect -0.073 0.029 -2.540 0.011
Odds ratios for other effects for binary Y
Pure natural IE 0.860 0.049 17.540 0.000
Tot natural DE 0.714 0.132 5.408 0.000
Total effect 0.614 0.120 5.138 0.000
CONFIDENCE INTERVALS OF MODEL RESULTS
Lower .5% Lower 2.5% Lower 5% Estimate Upper 5% Upper 2.5% Upper .5%
CIGUSE ON
INTENT 0.810 0.864 0.892 1.038 1.205 1.238 1.305
TX -0.903 -0.781 -0.715 -0.377 -0.044 0.019 0.145
INTENT ON
TX -0.327 -0.285 -0.266 -0.164 -0.065 -0.045 -0.009
Intercepts
INTENT 1.426 1.455 1.470 1.550 1.633 1.649 1.683
Thresholds
CIGUSE$1 3.056 3.056 3.056 3.056 3.056 3.056 3.056
Residual Variances
INTENT 0.634 0.668 0.684 0.776 0.864 0.883 0.918
New/Additional Parameters
INDIRECT 0.703 0.737 0.754 0.843 0.936 0.954 0.990
DIRECT 0.405 0.458 0.489 0.686 0.957 1.020 1.156
CONFIDENCE 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 -0.353 -0.306 -0.282 -0.171 -0.067 -0.047 -0.010
Direct effect -0.903 -0.781 -0.715 -0.377 -0.044 0.019 0.145
CONFIDENCE 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 -0.044 -0.038 -0.035 -0.021 -0.008 -0.006 -0.001
Pure natural DE -0.126 -0.109 -0.100 -0.053 -0.006 0.003 0.021
Total effect -0.149 -0.132 -0.121 -0.073 -0.026 -0.017 0.001
Odds ratios for binary Y
Tot natural IE 0.733 0.764 0.779 0.858 0.942 0.959 0.991
Pure natural DE 0.449 0.501 0.532 0.716 0.962 1.018 1.137
Total effect 0.374 0.422 0.448 0.614 0.840 0.893 1.008
Other effects
Pure natural IE -0.053 -0.045 -0.042 -0.025 -0.010 -0.007 -0.001
Tot natural DE -0.116 -0.100 -0.091 -0.048 -0.005 0.002 0.019
Total effect -0.149 -0.132 -0.121 -0.073 -0.026 -0.017 0.001
Odds ratios for other effects for binary Y
Pure natural IE 0.737 0.768 0.782 0.860 0.943 0.960 0.991
Tot natural DE 0.446 0.499 0.530 0.714 0.961 1.018 1.137
Total effect 0.374 0.422 0.448 0.614 0.840 0.893 1.008
TECHNICAL 1 OUTPUT
PARAMETER SPECIFICATION
TAU
CIGUSE$1
________
1 6
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 1 0
BETA
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0 2 3
INTENT 0 0 4
TX 0 0 0
PSI
CIGUSE INTENT TX
________ ________ ________
CIGUSE 0
INTENT 0 5
TX 0 0 0
PARAMETER SPECIFICATION FOR THE ADDITIONAL PARAMETERS
NEW/ADDITIONAL PARAMETERS
INDIRECT DIRECT
________ ________
1 7 8
STARTING VALUES
TAU
CIGUSE$1
________
1 1.513
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 1.456 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 0.391
TX 0.000 0.000 0.123
STARTING VALUES FOR THE ADDITIONAL PARAMETERS
NEW/ADDITIONAL PARAMETERS
INDIRECT DIRECT
________ ________
1 0.500 0.500
TECHNICAL 8 OUTPUT
E STEP ITER LOGLIKELIHOOD ABS CHANGE REL CHANGE ALGORITHM
1 -0.16607402D+04 0.0000000 0.0000000 EM
2 -0.14614378D+04 199.3024074 0.1200082 EM
3 -0.14524289D+04 9.0088645 0.0061644 EM
4 -0.14521743D+04 0.2546428 0.0001753 EM
5 -0.14521742D+04 0.0000772 0.0000001 EM
PLOT INFORMATION
The following plots are available:
Histograms (sample values)
Scatterplots (sample values)
Sample proportions
Bootstrap distributions
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\ex8.5.dgm
Beginning Time: 17:03:38
Ending Time: 17:04:04
Elapsed Time: 00:00:26
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