|
|
|
||||
|
|
|||||
Table 11.3 on page 386 using firstsex_wide.dat.
Model A:
Title:
Model A
Data:
File is d:\alda\firstsex_wide.dat ;
Variable:
Names are
id event7 event8 event9 event10 event11 event12 pt pas;
Missing are all (-9999) ;
usev are event7-event12;
classes =c(1);
categorical are event7-event12;
Analysis:
Type = mixture missing;
model:
%overall%
%c#1%
[event7$1*0 event8$1*0 event9$1*0 event10$1*0
event11$1*0 event12$1*0];
Notice that the deviance is -2*Loglikelihood. That is
-2*(-325.978) = 651.956. The threshold estimates in the output are just negative
of the parameter estimates in the book.
TESTS OF MODEL FIT
Loglikelihood
H0 Value -325.978
Information Criteria
Number of Free Parameters 6
Akaike (AIC) 663.955
Bayesian (BIC) 683.113
Sample-Size Adjusted BIC 664.111
(n* = (n + 2) / 24)
Chi-Square Test of Model Fit for the Latent Class Indicator Model Part
Pearson Chi-Square
Value 0.000
Degrees of Freedom 57
P-Value 1.0000
Likelihood Ratio Chi-Square
Value 0.000
Degrees of Freedom 57
P-Value 1.0000
Chi-Square Test for MCAR under the Unrestricted Latent Class Indicator Model
Pearson Chi-Square
Value 687.133
Degrees of Freedom 57
P-Value 0.0000
Likelihood Ratio Chi-Square
Value 551.062
Degrees of Freedom 57
P-Value 0.0000
FINAL CLASS COUNTS AND PROPORTIONS OF TOTAL SAMPLE SIZE BASED ON ESTIMATED POSTERIOR PROBABILITIES
Class 1 180.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY CLASS MEMBERSHIP
Class Counts and Proportions
Class 1 180 1.00000
Average Class Probabilities by Class
1
Class 1 1.000
MODEL RESULTS
Estimates S.E. Est./S.E.
CLASS 1
LATENT CLASS INDICATOR MODEL PART
Class 1
Thresholds
EVENT7$1 2.398 0.270 8.892
EVENT8$1 3.117 0.386 8.069
EVENT9$1 1.720 0.222 7.759
EVENT10$1 1.287 0.210 6.133
EVENT11$1 1.163 0.229 5.076
EVENT12$1 0.731 0.239 3.062
LATENT CLASS INDICATOR MODEL PART IN PROBABILITY SCALE
Class 1
EVENT7
Category 1 0.917 0.021 44.497
Category 2 0.083 0.021 4.045
EVENT8
Category 1 0.958 0.016 61.027
Category 2 0.042 0.016 2.704
EVENT9
Category 1 0.848 0.029 29.701
Category 2 0.152 0.029 5.320
EVENT10
Category 1 0.784 0.036 22.027
Category 2 0.216 0.036 6.084
EVENT11
Category 1 0.762 0.042 18.330
Category 2 0.238 0.042 5.728
EVENT12
Category 1 0.675 0.052 12.890
Category 2 0.325 0.052 6.206
Model B:
Title:
Model B
Data:
File is d:\alda\firstsex_wide.dat ;
Variable:
Names are
id event7 event8 event9 event10 event11 event12 pt pas;
Missing are all (-9999) ;
usev are event7-event12 pt;
classes =c(1);
categorical are event7-event12;
Analysis:
Type = mixture missing;
model:
%overall% event7 - event12 on pt (1); %c#1% [event7$1*0 event8$1*0 event9$1*0 event10$1*0 event11$1*0 event12$1*0];
The calculation of deviance is the same as shown in the previous example: -2*(-317.331) = 634.662.
TESTS OF MODEL FIT
Loglikelihood
H0 Value -317.331
Information Criteria
Number of Free Parameters 7
Akaike (AIC) 648.662
Bayesian (BIC) 671.012
Sample-Size Adjusted BIC 648.843
(n* = (n + 2) / 24)
FINAL CLASS COUNTS AND PROPORTIONS OF TOTAL SAMPLE SIZE BASED ON ESTIMATED POSTERIOR PROBABILITIES
Class 1 180.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY CLASS MEMBERSHIP
Class Counts and Proportions
Class 1 180 1.00000
Average Class Probabilities by Class
1
Class 1 1.000
MODEL RESULTS
Estimates S.E. Est./S.E.
CLASS 1
LATENT CLASS INDICATOR MODEL PART
Class 1
Thresholds
EVENT7$1 2.994 0.305 9.832
EVENT8$1 3.700 0.424 8.726
EVENT9$1 2.281 0.281 8.110
EVENT10$1 1.823 0.258 7.078
EVENT11$1 1.654 0.276 5.997
EVENT12$1 1.179 0.270 4.372
EVENT7 ON
PT 0.874 0.217 4.024
EVENT8 ON
PT 0.874 0.217 4.024
EVENT9 ON
PT 0.874 0.217 4.024
EVENT10 ON
PT 0.874 0.217 4.024
EVENT11 ON
PT 0.874 0.217 4.024
EVENT12 ON
PT 0.874 0.217 4.024
Model C:
Title:
Model C
Data:
File is d:\firstsex_wide.dat ;
Variable:
Names are
id event7 event8 event9 event10 event11 event12 pt pas;
Missing are all (-9999) ;
usev are event7-event12 pas;
classes =c(1);
categorical are event7-event12;
Analysis:
Type = mixture missing;
model:
%overall% event7 - event12 on pas (1); %c#1% [event7$1*0 event8$1*0 event9$1*0 event10$1*0 event11$1*0 event12$1*0];
The deviance is -2*(-318.584) = 637.168.
TESTS OF MODEL FIT
Loglikelihood
H0 Value -318.584
Information Criteria
Number of Free Parameters 7
Akaike (AIC) 651.169
Bayesian (BIC) 673.519
Sample-Size Adjusted BIC 651.350
(n* = (n + 2) / 24)
FINAL CLASS COUNTS AND PROPORTIONS OF TOTAL SAMPLE SIZE BASED ON ESTIMATED POSTERIOR PROBABILITIES
Class 1 180.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY CLASS MEMBERSHIP
Class Counts and Proportions
Class 1 180 1.00000
Average Class Probabilities by Class
1
Class 1 1.000
MODEL RESULTS
Estimates S.E. Est./S.E.
CLASS 1
LATENT CLASS INDICATOR MODEL PART
Class 1
Thresholds
EVENT7$1 2.465 0.266 9.262
EVENT8$1 3.159 0.386 8.176
EVENT9$1 1.730 0.230 7.516
EVENT10$1 1.285 0.214 5.993
EVENT11$1 1.136 0.230 4.934
EVENT12$1 0.642 0.239 2.687
EVENT7 ON
PAS 0.443 0.109 4.066
EVENT8 ON
PAS 0.443 0.109 4.066
EVENT9 ON
PAS 0.443 0.109 4.066
EVENT10 ON
PAS 0.443 0.109 4.066
EVENT11 ON
PAS 0.443 0.109 4.066
EVENT12 ON
PAS 0.443 0.109 4.066
Model D:
Title:
Model D
Data:
File is d:\alda\firstsex_wide.dat ;
Variable:
Names are
id event7 event8 event9 event10 event11 event12 pt pas;
Missing are all (-9999) ;
usev are event7-event12 pt pas;
classes =c(1);
categorical are event7-event12;
Analysis:
Type = mixture missing;
model:
%overall% event7 - event12 on pt (1); event7 - event12 on pas (2); %c#1% [event7$1*0 event8$1*0 event9$1*0 event10$1*0 event11$1*0 event12$1*0];
The deviance is -2*(-314.573) = 629.146.
TESTS OF MODEL FIT
Loglikelihood
H0 Value -314.573
Information Criteria
Number of Free Parameters 8
Akaike (AIC) 645.147
Bayesian (BIC) 670.691
Sample-Size Adjusted BIC 645.355
(n* = (n + 2) / 24)
FINAL CLASS COUNTS AND PROPORTIONS OF TOTAL SAMPLE SIZE BASED ON ESTIMATED POSTERIOR PROBABILITIES
Class 1 180.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY CLASS MEMBERSHIP
Class Counts and Proportions
Class 1 180 1.00000
Average Class Probabilities by Class
1
Class 1 1.000
MODEL RESULTS
Estimates S.E. Est./S.E.
CLASS 1
LATENT CLASS INDICATOR MODEL PART
Class 1
Thresholds
EVENT7$1 2.893 0.309 9.359
EVENT8$1 3.585 0.427 8.397
EVENT9$1 2.150 0.284 7.562
EVENT10$1 1.693 0.265 6.398
EVENT11$1 1.518 0.281 5.398
EVENT12$1 1.010 0.275 3.669
EVENT7 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
EVENT8 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
EVENT9 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
EVENT10 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
EVENT11 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
EVENT12 ON
PT 0.661 0.237 2.788
PAS 0.296 0.120 2.479
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services