Normal Theory Maximum Likelihood Estimation
This is the Mplus
code deployed by
FitModelML()
. Note that FitModelMI()
used
FitModelML()
to fit the model on each of the imputed data
sets.
DATA:
FILE = data.dat;
VARIABLE:
NAMES = X M Y;
USEVARIABLES = X M Y;
MISSING = ALL (-999);
ANALYSIS:
TYPE = GENERAL;
ESTIMATOR = ML;
MODEL:
Y ON X;
Y ON M (B);
M ON X (A);
X WITH X;
SAVEDATA:
RESULTS ARE results.out;
TECH3 IS tech3.out;
Nonparametric Bootstrap Confidence Intervals
This is the Mplus
code deployed by
NBML()
.
DATA:
FILE = data.dat;
VARIABLE:
NAMES = X M Y;
USEVARIABLES = X M Y;
MISSING = ALL (-999);
ANALYSIS:
TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 5000;
MODEL:
Y ON X;
Y ON M (B);
M ON X (A);
X WITH X;
MODEL CONSTRAINT:
NEW(AB);
AB = A*B;
OUTPUT:
CINT(bootstrap); ! percentile
! CINT(bcbootstrap); ! bias-corrected
Multiple Imputation
This is the Mplus
code deployed by
ImputeData()
.
DATA:
FILE = data.dat;
VARIABLE:
NAMES = X M Y;
USEVARIABLES = X M Y;
MISSING = ALL (-999);
ANALYSIS:
TYPE = BASIC;
DATA IMPUTATION:
IMPUTE = X M Y;
NDATASETS = 100;
SAVE = mi*.dat;