Mplus VERSION 6
MUTHEN & MUTHEN
04/25/2010 11:01 PM
INPUT INSTRUCTIONS
Title: app13
Recid2
Recidivism Example
Discrete time survival analysis in mixture modeling framework
Event is first arrest after prison release
Time scale is 4 week intervals, numbered 1 - 13
Hazard function is unstructured
Finaid included with time-invariant effects --> proportionality assumed
One-class model
VARIABLES:
u1-u13 = I(first arrest, months 1-13), censored=999
finaid = 1 if financial aid was provided, 0 if no aid (INTERVENTION)
age = age at release (in years)
race = 1 if black, 0 if white
wexp = 1 if prior work experience, 0 if no work experience
mar = 1 if married, 0 if unmarried
parole = 1 if paroled, 0 if not paroled
priors = number of prior arrests
educ = years of schooling
empb1-empb13 = I(one of more weeks of employment during interval)
tr1-tr2 = Training data for long-term survivor class
Data:
File is recid.dat;
Variable:
Names are id u1-u13 finaid age race wexp mar parole
priors educ empb1-empb13 tr1 tr2;
Missing are all (999);
Usevariables are u1-u13 finaid;
Categorical are u1-u13;
Classes = c(1);
Analysis:
Type = Mixture Missing;
MIterations = 1000;
MConvergence = 0.000001;
LogCriterion = 0.0000001;
Convergence = 0.000001;
Model:
%overall%
f by u1-u13@1;
f on finaid;
[f@0];
%c#1%
[u1$1*4.7 u2$1*4.0 u3$1*4.1 u4$1*3.9 u5$1*3.4 u6$1*3.9 u7$1*3.6];
[u8$1*4.3 u9$1*3.5 u10$1*3.5 u11$1*3.7 u12$1*3.6 u13$1*3.3];
f on finaid*;
[f@0];
Output:
Tech1;
Tech8;
*** WARNING in ANALYSIS command
Starting with Version 5, TYPE=MISSING is the default for all analyses.
To obtain listwise deletion, use LISTWISE=ON in the DATA command.
1 WARNING(S) FOUND IN THE INPUT INSTRUCTIONS
app13
Recid2
Recidivism Example
Discrete time survival analysis in mixture modeling framework
Event is first arrest after prison release
Time scale is 4 week intervals, numbered 1 - 13
Hazard function is unstructured
Finaid included with time-invariant effects --> proportionality assumed
One-class model
VARIABLES:
u1-u13 = I(first arrest, months 1-13), censored=999
finaid = 1 if financial aid was provided, 0 if no aid (INTERVENTION)
age = age at release (in years)
race = 1 if black, 0 if white
wexp = 1 if prior work experience, 0 if no work experience
mar = 1 if married, 0 if unmarried
parole = 1 if paroled, 0 if not paroled
priors = number of prior arrests
educ = years of schooling
empb1-empb13 = I(one of more weeks of employment during interval)
tr1-tr2 = Training data for long-term survivor class
SUMMARY OF ANALYSIS
Number of groups 1
Number of observations 432
Number of dependent variables 13
Number of independent variables 1
Number of continuous latent variables 1
Number of categorical latent variables 1
Observed dependent variables
Binary and ordered categorical (ordinal)
U1 U2 U3 U4 U5 U6
U7 U8 U9 U10 U11 U12
U13
Observed independent variables
FINAID
Continuous latent variables
F
Categorical latent variables
C
Estimator MLR
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 1000
Convergence criteria
Loglikelihood change 0.100D-06
Relative loglikelihood change 0.100D-06
Derivative 0.100D-05
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-05
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-05
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
Maximum number of iterations for H1 2000
Convergence criterion for H1 0.100D-03
Optimization algorithm EMA
Link LOGIT
Input data file(s)
recid.dat
Input data format FREE
SUMMARY OF DATA
Number of missing data patterns 13
Number of y missing data patterns 0
Number of u missing data patterns 13
COVARIANCE COVERAGE OF DATA
Minimum covariance coverage value 0.100
PROPORTION OF DATA PRESENT FOR U
Covariance Coverage
U1 U2 U3 U4 U5
________ ________ ________ ________ ________
U1 1.000
U2 0.991 0.991
U3 0.972 0.972 0.972
U4 0.956 0.956 0.956 0.956
U5 0.938 0.938 0.938 0.938 0.938
U6 0.907 0.907 0.907 0.907 0.907
U7 0.889 0.889 0.889 0.889 0.889
U8 0.866 0.866 0.866 0.866 0.866
U9 0.854 0.854 0.854 0.854 0.854
U10 0.829 0.829 0.829 0.829 0.829
U11 0.803 0.803 0.803 0.803 0.803
U12 0.785 0.785 0.785 0.785 0.785
U13 0.764 0.764 0.764 0.764 0.764
Covariance Coverage
U6 U7 U8 U9 U10
________ ________ ________ ________ ________
U6 0.907
U7 0.889 0.889
U8 0.866 0.866 0.866
U9 0.854 0.854 0.854 0.854
U10 0.829 0.829 0.829 0.829 0.829
U11 0.803 0.803 0.803 0.803 0.803
U12 0.785 0.785 0.785 0.785 0.785
U13 0.764 0.764 0.764 0.764 0.764
Covariance Coverage
U11 U12 U13
________ ________ ________
U11 0.803
U12 0.785 0.785
U13 0.764 0.764 0.764
UNIVARIATE PROPORTIONS AND COUNTS FOR CATEGORICAL VARIABLES
U1
Category 1 0.991 428.000
Category 2 0.009 4.000
U2
Category 1 0.981 420.000
Category 2 0.019 8.000
U3
Category 1 0.983 413.000
Category 2 0.017 7.000
U4
Category 1 0.981 405.000
Category 2 0.019 8.000
U5
Category 1 0.968 392.000
Category 2 0.032 13.000
U6
Category 1 0.980 384.000
Category 2 0.020 8.000
U7
Category 1 0.974 374.000
Category 2 0.026 10.000
U8
Category 1 0.987 369.000
Category 2 0.013 5.000
U9
Category 1 0.970 358.000
Category 2 0.030 11.000
U10
Category 1 0.969 347.000
Category 2 0.031 11.000
U11
Category 1 0.977 339.000
Category 2 0.023 8.000
U12
Category 1 0.973 330.000
Category 2 0.027 9.000
U13
Category 1 0.964 318.000
Category 2 0.036 12.000
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -534.835
H0 Scaling Correction Factor 1.000
for MLR
Information Criteria
Number of Free Parameters 14
Akaike (AIC) 1097.670
Bayesian (BIC) 1154.628
Sample-Size Adjusted BIC 1110.200
(n* = (n + 2) / 24)
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 432.00000 1.00000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 432.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 432 1.00000
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1
1 1.000
MODEL RESULTS
Two-Tailed
Estimate S.E. Est./S.E. P-Value
Latent Class 1
F BY
U1 1.000 0.000 999.000 999.000
U2 1.000 0.000 999.000 999.000
U3 1.000 0.000 999.000 999.000
U4 1.000 0.000 999.000 999.000
U5 1.000 0.000 999.000 999.000
U6 1.000 0.000 999.000 999.000
U7 1.000 0.000 999.000 999.000
U8 1.000 0.000 999.000 999.000
U9 1.000 0.000 999.000 999.000
U10 1.000 0.000 999.000 999.000
U11 1.000 0.000 999.000 999.000
U12 1.000 0.000 999.000 999.000
U13 1.000 0.000 999.000 999.000
F ON
FINAID -0.374 0.192 -1.948 0.051
Intercepts
F 0.000 0.000 999.000 999.000
Thresholds
U1$1 4.503 0.520 8.662 0.000
U2$1 3.789 0.363 10.450 0.000
U3$1 3.906 0.379 10.316 0.000
U4$1 3.754 0.362 10.360 0.000
U5$1 3.235 0.291 11.135 0.000
U6$1 3.700 0.367 10.079 0.000
U7$1 3.449 0.346 9.978 0.000
U8$1 4.124 0.459 8.992 0.000
U9$1 3.305 0.306 10.809 0.000
U10$1 3.276 0.319 10.261 0.000
U11$1 3.569 0.373 9.578 0.000
U12$1 3.422 0.344 9.934 0.000
U13$1 3.097 0.307 10.102 0.000
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.714E-01
(ratio of smallest to largest eigenvalue)
TECHNICAL 1 OUTPUT
PARAMETER SPECIFICATION FOR LATENT CLASS 1
NU
FINAID
________
1 0
LAMBDA
FINAID
________
FINAID 0
THETA
FINAID
________
FINAID 0
ALPHA
FINAID
________
1 0
BETA
FINAID
________
FINAID 0
PSI
FINAID
________
FINAID 0
PARAMETER SPECIFICATION FOR LATENT CLASS INDICATOR MODEL PART
TAU(U) FOR LATENT CLASS 1
U1$1 U2$1 U3$1 U4$1 U5$1
________ ________ ________ ________ ________
1 1 2 3 4 5
TAU(U) FOR LATENT CLASS 1
U6$1 U7$1 U8$1 U9$1 U10$1
________ ________ ________ ________ ________
1 6 7 8 9 10
TAU(U) FOR LATENT CLASS 1
U11$1 U12$1 U13$1
________ ________ ________
1 11 12 13
PARAMETER SPECIFICATION FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
C#1
________
1 0
GAMMA(C)
FINAID
________
C#1 0
PARAMETER SPECIFICATION FOR LATENT CLASS INDICATOR GROWTH MODEL PART
LAMBDA(F) FOR LATENT CLASS 1
F
________
U1 0
U2 0
U3 0
U4 0
U5 0
U6 0
U7 0
U8 0
U9 0
U10 0
U11 0
U12 0
U13 0
ALPHA(F) FOR LATENT CLASS 1
F
________
1 0
GAMMA(F) FOR LATENT CLASS 1
FINAID
________
F 14
STARTING VALUES FOR LATENT CLASS 1
NU
FINAID
________
1 0.000
LAMBDA
FINAID
________
FINAID 1.000
THETA
FINAID
________
FINAID 0.000
ALPHA
FINAID
________
1 0.000
BETA
FINAID
________
FINAID 0.000
PSI
FINAID
________
FINAID 0.125
STARTING VALUES FOR LATENT CLASS INDICATOR MODEL PART
TAU(U) FOR LATENT CLASS 1
U1$1 U2$1 U3$1 U4$1 U5$1
________ ________ ________ ________ ________
1 4.700 4.000 4.100 3.900 3.400
TAU(U) FOR LATENT CLASS 1
U6$1 U7$1 U8$1 U9$1 U10$1
________ ________ ________ ________ ________
1 3.900 3.600 4.300 3.500 3.500
TAU(U) FOR LATENT CLASS 1
U11$1 U12$1 U13$1
________ ________ ________
1 3.700 3.600 3.300
STARTING VALUES FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
C#1
________
1 0.000
GAMMA(C)
FINAID
________
C#1 0.000
STARTING VALUES FOR LATENT CLASS INDICATOR GROWTH MODEL PART
LAMBDA(F) FOR CLASS LATENT CLASS 1
F
________
U1 1.000
U2 1.000
U3 1.000
U4 1.000
U5 1.000
U6 1.000
U7 1.000
U8 1.000
U9 1.000
U10 1.000
U11 1.000
U12 1.000
U13 1.000
ALPHA(F) FOR LATENT CLASS 1
F
________
1 0.000
GAMMA(F) FOR LATENT CLASS 1
FINAID
________
F 0.000
TECHNICAL 8 OUTPUT
ITER LOGLIKELIHOOD ABS CHANGE REL CHANGE CLASS COUNTS ALGORITHM
1 -0.53679672D+03 0.0000000 0.0000000 432.000 EM
2 -0.53485113D+03 1.9455911 0.0036244 432.000 EM
3 -0.53483514D+03 0.0159898 0.0000299 432.000 EM
4 -0.53483514D+03 0.0000010 0.0000000 432.000 EM
5 -0.53483514D+03 0.0000000 0.0000000 432.000 EM
Beginning Time: 23:01:08
Ending Time: 23:01:08
Elapsed Time: 00:00:00
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) 1998-2010 Muthen & Muthen
Back to table of examples