mi (version 0.10-2)

mi.scatterplot: Multiple Imputation Scatterplot

Description

A function for plotting observed and imputed values for a variable .

Usage

mi.scatterplot( Yobs, Yimp, X = NULL, xlab = NULL, ylab = NULL, 
                            main = "Imputed Variable Scatter Plot", 
                             display.zero = TRUE, gray.scale = FALSE, 
                              obs.col = rgb( 0, 0, 1 ), 
                              imp.col = rgb( 1, 0, 0 ), 
                              obs.pch = 20 , imp.pch = 20, 
                              obs.cex = 0.3, imp.cex = 0.3, 
                              obs.lty = 1  , imp.lty = 1, 
                              obs.lwd = 2.5, imp.lwd = 2.5, ... )
marginal.scatterplot ( data, object, use.imputed.X = FALSE, ...  )

Arguments

Yobs

observed values.

Yimp

imputed values.

X

variable to plot on the x axis.

xlab

label on the x axis.

ylab

label on the y axis.

display.zero

if set to FALSE zeros will not be displayed. Default is TRUE.

main

main title of the plot.

gray.scale

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

obs.col

color for the observed variable. Default is "blue".

imp.col

color for the imputed variable. Default is "red".

obs.pch

data symbol for observed variable. Default is 20.

imp.pch

data symbol for imputed variable. Default is 20.

obs.cex

text size for observed variable. Default is 0.3.

imp.cex

text size for imputed variable. Default is 0.3.

obs.lty

line type for observed variable. Default is 1.

imp.lty

line type for imputed variable. Default is 1.

obs.lwd

line width for observed variable. Default is 2.5.

imp.lwd

line width for imputed variable. Default is 2.5.

...

Other options for 'plot' function.

data

missing data.

object

mi object.

use.imputed.X

If you want to use the imputed X. Default is FALSE.

Value

• A scatterplot with the observed and the imputed values plotted against a chosen variable.

Details

Since several data points can have the same data values, especially in discrete variables, small random number is added to each value so that points do not fall on top of each other. See help on jitter for more details. Lowess line is fitted to both imputed and observed data.

References

Yu-Sung Su, Andrew Gelman, Jennifer Hill, Masanao Yajima. (2011). Multiple Imputation with Diagnostics (mi) in R: Opening Windows into the Black Box. Journal of Statistical Software 45(2). Kobi Abayomi, Andrew Gelman and Marc Levy. (2008). Diagnostics for multivariate imputations. Applied Statistics 57, Part 3: 273--291. Andrew Gelman and Jennifer Hill. (2007). Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press.

See Also

mi, plot

Examples

# true data
  x<-rnorm(100,0,1) # N(0,1)
  y<-rnorm(100,(1+2*x),1.2) # y ~ 1 + 2*x + N(0,1.2)
  # create artificial missingness on y
  y[seq(1,100,10)]<-NA
  dat.xy <- data.frame(x,y)
  # imputation
  imp.cont<-mi.continuous(y~x, data = dat.xy)
  mi.scatterplot(y,imputed(imp.cont,y))