mi (version 0.06-5)

plot.mi: Diagnostic Plots for multiple imputation object

Description

Diagnostic plots for testing the fit of the imputation method to the observed data.

Usage

## S3 method for class 'mi,missing':
plot( x, m = 1, vrb = NULL, vrb.name = "Variable Score",
                    gray.scale = FALSE, mfrow=c( 1, 4 ), ...)

Arguments

x

mi object generated by the mi function.

m

The m-th imputation. By default is 1.

vrb

A chosen variable for the scatter plot.

vrb.name

A name of the vrb variable.

gray.scale

When set to TRUE, makes the plot into gray scale with predefined color and line type.

mfrow

See par for details.

...

Arguments for other methods, not used.

Value

• Histograms, scatterplots, and residual plots of the fit of the imputation models. Binned residual plots are for each dichotomous variable.

Details

For each variable, observed values are in blue, the imputed values are in red. In the scatterplot the observed and the imputed are plotted versus a variable the users can choose. By default the values are plotted against an index number but it strongly recommended to use a variable containing more information. Fitted lowess lines are also plotted for both observed and imputed data. A small amount of random noise (jittering) is added to the points so that they do not fall on top of each other.

References

Kobi Abayomi, Andrew Gelman and Marc Levy. (2008). Diagnostics for multivariate imputations. Applied Statistics 57, Part 3: 273--291. Andrew Gelman and Maria Grazia Pittau. A flexible program for missing-data imputation and model checking. Technical report. Columbia University, New York. Andrew Gelman and Jennifer Hill. (2007). Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press.

See Also

mi, mi.scatterplot, mi.hist

Examples

data(CHAIN)
  imp.CHAIN <- mi(CHAIN, n.iter=6, preprocess=FALSE, add.priors=FALSE)
  plot(imp.CHAIN)