Imputation of Counterfactual
imputeCounter imputes counterfactual missing values for a gjrm model object.
imputeCounter(x, m = 10, nm.end)
x |
A fitted |
m |
Number of imputed response vectors. |
nm.end |
Name endogenous variable. |
This function generates m sets of imputed values for the outcome of interest under a fitted joint causal model. The algorithm draws parameters from the posterior distribution of the model which are then used to obtain simulated responses (from the posterior predictive distribution of the missing values) via a rejection algorithm. The bound for acceptance/rejection is obtained via a trust region optimisation.
The imputed values are used to create m complete imputed datasets and perform complete data analysis and inference about the parameters of interest using any summary statistics.
It returns a list containing m imputed response vectors.
Maintainer: Giampiero Marra giampiero.marra@ucl.ac.uk
Robert C. and Casella G. (2004). Monte Carlo Statistical Methods. New York: Springer-Verlag.
Ripley B. D. (1987) Stochastic Simulation. New York: John Wiley & Sons, Inc.
## see examples for gjrm
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.