runMI: Multiply impute and analyze data using lavaan
Description
This function takes data with missing observations, multiple imputes the data, runs a SEM using lavaan and combines the results using Rubin's rules.Usage
runMI(data.mat, data.model, m, miPackage="Amelia", digits = 3, ...)
Arguments
data.mat
Data frame with missing observations.
data.model
lavaan syntax for the the model to be analyzed.
m
Number of imputations wanted.
miPackage
Package to be used for imputation. Currently runMI only uses Amelia or mice for imputation.
digits
Number of digits to print in the results.
...
Other arguments to be passed to the imputation package
Value
- runMI returns a list with pooled fit indices, estimates, standard errors and fraction missing information
- fitPooled fit information. The first set of fit information are simply averaged across imputations and are not trustworthy. The second set of fit information, is a pooled Chi-square statistic based on Li, Meng, Raghunathan, & Rubin (1991)
- parametersPooled parameter estimates and standard errors. Wald statistics and p values are computed from the pooled estimates and standard errors. Also contains two estimates of Fraction of Missing Information (FMI). Includes asymptotic FMI (FMI.1) and FMI that is corrected for small numbers of imputation (FMI.2)
References
Li, K.H., Meng, X.-L., Raghunathan, T.E. and Rubin, D.B. (1991). Significance Levels From Repeated p-values with Multiply-Imputed Data. Statistica Sinica, 1, 65-92.
Rubin, D.B. (1987) Multiple Imputation for Nonresponse in Surveys. J. Wiley & Sons, New York.