Package 'ggmice'

Box-and-whisker plot of observed and imputed data

Description

Box-and-whisker plot of observed and imputed data

Usage

bwplot(...)

Arguments

...	Any arguments passed to the function.

Value

The output of mice::bwplot and a message about the ggmice equivalent.

Examples

imp <- mice::mice(mice::nhanes, maxit = 1, printFlag = FALSE)
bwplot(imp)

---

Densityplot of observed and imputed data

Description

Densityplot of observed and imputed data

Usage

densityplot(...)

Arguments

...	Any arguments passed to the function.

Value

The output of mice::densityplot and a message about the ggmice equivalent.

Examples

imp <- mice::mice(mice::nhanes, maxit = 1, printFlag = FALSE)
densityplot(imp)

---

Plot incomplete or imputed data

Description

Plot incomplete or imputed data

Usage

ggmice(data = NULL, mapping = ggplot2::aes())

Arguments

data	An incomplete dataset (of class data.frame), or an object of class mice::mids.
mapping	A list of aesthetic mappings created with ggplot2::aes().

Value

An object of class ggplot2::ggplot. The ggmice function returns output equivalent to ggplot2::ggplot output, with a few important exceptions:

• The theme is set to theme_mice.

• The color scale is set to the mice::mdc colors.

• The colour aesthetic is set to .where, an internally defined variable which distinguishes observed data from missing data or imputed data (for incomplete and imputed data, respectively).

See Also

See the ggmice vignette to use the ggmice() function on incomplete data or imputed data.

Examples

dat <- mice::nhanes
ggmice(dat, ggplot2::aes(x = age, y = bmi)) + ggplot2::geom_point()
imp <- mice::mice(dat, print = FALSE)
ggmice(imp, ggplot2::aes(x = age, y = bmi)) + ggplot2::geom_point()

---

Plot correlations between (incomplete) variables

Description

Plot correlations between (incomplete) variables

Usage

plot_corr(
  data,
  vrb = "all",
  label = FALSE,
  square = TRUE,
  diagonal = FALSE,
  rotate = FALSE,
  caption = TRUE
)

Arguments

data	A dataset of class data.frame, tibble, or matrix.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
label	Logical indicating whether correlation values should be displayed.
square	Logical indicating whether the plot tiles should be squares.
diagonal	Logical indicating whether the correlation of each variable with itself should be displayed.
rotate	Logical indicating whether the variable name labels should be rotated 90 degrees.
caption	Logical indicating whether the figure caption should be displayed.

Value

An object of class ggplot2::ggplot.

Examples

# plot correlations for all columns
plot_corr(mice::nhanes)

# plot correlations for specific columns by supplying a character vector
plot_corr(mice::nhanes, c("chl", "hyp"))

# plot correlations for specific columns by supplying unquoted variable names
plot_corr(mice::nhanes, c(chl, hyp))

# plot correlations for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_corr(mice::nhanes, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_corr(mice::nhanes, my_variables))

---

Plot the influx and outflux of a multivariate missing data pattern

Description

Plot the influx and outflux of a multivariate missing data pattern

Usage

plot_flux(data, vrb = "all", label = TRUE, caption = TRUE)

Arguments

data	An incomplete dataset of class data.frame or matrix.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
label	Logical indicating whether variable names should be displayed within the plot (the default) or with colors in the legend.
caption	Logical indicating whether the figure caption should be displayed.

Value

An object of class ggplot2::ggplot.

Examples

# plot flux for all columns
plot_flux(mice::nhanes)

# plot flux for specific columns by supplying a character vector
plot_flux(mice::nhanes, c("chl", "hyp"))

# plot flux for specific columns by supplying unquoted variable names
plot_flux(mice::nhanes, c(chl, hyp))

# plot flux for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_flux(mice::nhanes, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_flux(mice::nhanes, my_variables))

---

Plot missingness in a dataset

Description

Usage

plot_miss(
  data,
  vrb = "all",
  ordered = FALSE,
  rotate = FALSE,
  grid = FALSE,
  square = FALSE
)

Arguments

data	An incomplete dataset of class data.frame or matrix.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
ordered	Logical indicating whether rows should be ordered according to their pattern.
rotate	Logical indicating whether the variable name labels should be rotated 90 degrees.
grid	Logical indicating whether borders should be present between tiles.
square	Logical indicating whether the plot tiles should be squares, defaults to squares.

Value

An object of class ggplot2::ggplot.

Examples

# plot correlations for all columns
plot_miss(mice::nhanes)

# plot correlations for specific columns by supplying a character vector
plot_miss(mice::nhanes, c("chl", "hyp"))

# plot correlations for specific columns by supplying unquoted variable names
plot_miss(mice::nhanes, c(chl, hyp))

# plot correlations for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_miss(mice::nhanes, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_miss(mice::nhanes, my_variables))

# plot larger dataset
plot_miss(mice::boys)
plot_miss(mice::boys, ordered = TRUE)

---

Plot the missing data pattern of an incomplete dataset

Description

Plot the missing data pattern of an incomplete dataset

Usage

plot_pattern(
  data,
  vrb = "all",
  square = TRUE,
  rotate = FALSE,
  cluster = NULL,
  npat = NULL,
  caption = TRUE
)

Arguments

data	An incomplete dataset of class data.frame or matrix.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
square	Logical indicating whether the plot tiles should be squares, defaults to squares to mimic mice::md.pattern().
rotate	Logical indicating whether the variable name labels should be rotated 90 degrees.
cluster	Optional character string specifying which variable should be used for clustering (e.g., for multilevel data).
npat	Optional numeric input specifying the maximum number of missing data patterns to be visualized, defaults to all patterns.
caption	Logical indicating whether the figure caption should be displayed.

Value

An object of class ggplot2::ggplot.

Examples

# plot missing data pattern for all columns
plot_pattern(mice::nhanes)

# plot missing data pattern for specific columns by supplying a character vector
plot_pattern(mice::nhanes, c("chl", "hyp"))

# plot missing data pattern for specific columns by supplying unquoted variable names
plot_pattern(mice::nhanes, c(chl, hyp))

# plot missing data pattern for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_pattern(mice::nhanes, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_pattern(mice::nhanes, my_variables))

# plot missing data pattern by cluster
plot_pattern(mice::nhanes, cluster = "age")

---

Plot the predictor matrix of an imputation model

Description

Plot the predictor matrix of an imputation model

Usage

plot_pred(
  data,
  vrb = "all",
  method = NULL,
  label = TRUE,
  square = TRUE,
  rotate = FALSE
)

Arguments

data	A predictor matrix for mice, typically generated with mice::make.predictorMatrix or mice::quickpred, or an object of class mice::mids.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
method	Character string or vector with imputation methods.
label	Logical indicating whether predictor matrix values should be displayed.
square	Logical indicating whether the plot tiles should be squares.
rotate	Logical indicating whether the variable name labels should be rotated 90 degrees.

Details

The predictor matrix in mice::mice determines the role an imputation model predictor takes in the imputation model. The rows correspond to incomplete target variables, and the columns to imputation model predictors.

A value of 1 indicates that the column variable is a predictor to impute the target (row) variable. The value 0 means that it is not used as predictor.

Imputation methods for multilevel data use other codes than 0 and 1:

• Methods ⁠2l.bin⁠, ⁠2l.lmer⁠, ⁠2l.norm⁠, ⁠2l.pan⁠, ⁠2lonly.mean⁠, ⁠2lonly.norm⁠ and ⁠2lonly.pmm⁠ use code -2 to indicate the class variable;

• Methods ⁠2l.bin⁠, ⁠2l.lmer⁠, ⁠2l.norm⁠ and ⁠2l.pan⁠ use code 2 to indicate the random effects;

• Method ⁠2l.pan⁠ uses codes 3 and 4 to add class means to codes 1 and 2 respectively.

Value

An object of class ggplot2::ggplot.

References

van Buuren, S. (2018). Flexible imputation of missing data. Chapman and Hall/CRC. stefvanbuuren.name/fimd

Examples

# generate a predictor matrix
pred <- mice::quickpred(mice::nhanes)

# plot predictor matrix for all columns
plot_pred(pred)

# plot predictor matrix for specific columns by supplying a character vector
plot_pred(pred, c("chl", "hyp"))

# plot predictor matrix for specific columns by supplying unquoted variable names
plot_pred(pred, c(chl, hyp))

# plot predictor matrix for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_pred(pred, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_pred(pred, my_variables))

# plot predictor matrix of mids object
imp <- mice::mice(mice::nhanes, print = FALSE)
plot_pred(imp)

---

Plot the trace lines of the imputation algorithm

Description

Plot the trace lines of the imputation algorithm

Usage

plot_trace(data, vrb = "all", trend = FALSE, legend = TRUE)

Arguments

data	An object of class mice::mids.
vrb	String, vector, or unquoted expression with variable name(s), default is "all".
trend	Logical indicating whether a smoothened trend should be added, default is FALSE.
legend	Logical indicating whether the plot legend should be visible, default is TRUE.

Details

The vrb argument is "quoted" via rlang::enexpr() and evaluated according to tidy evaluation principles. In practice, this technical nuance only affects users when passing an object from the environment (e.g., a vector of variable names) to the vrb argument. In such cases, the object must be "unquoted" via the ⁠!!⁠ prefix operator.

Value

An object of class ggplot2::ggplot.

Examples

# create [mice::mids] object with [mice::mice()]
imp <- mice::mice(mice::nhanes, print = FALSE)

# plot trace lines for all imputed columns
plot_trace(imp)

# plot trace lines for specific columns by supplying a string or character vector
plot_trace(imp, "chl")
plot_trace(imp, c("chl", "hyp"))
# plot trace lines for specific columns by supplying unquoted variable names
plot_trace(imp, chl)
plot_trace(imp, c(chl, hyp))

# plot trace lines for specific columns by passing an object with variable names
# from the environment, unquoted with `!!`
my_variables <- c("chl", "hyp")
plot_trace(imp, !!my_variables)
# object with variable names must be unquoted with `!!`
try(plot_trace(imp, my_variables))

---

Stripplot of observed and imputed data

Description

Stripplot of observed and imputed data

Usage

stripplot(...)

Arguments

...	Any arguments passed to the function.

Value

The output of mice::stripplot and a message about the ggmice equivalent.

Examples

imp <- mice::mice(mice::nhanes, maxit = 1, printFlag = FALSE)
stripplot(imp)

---

Scatterplot of observed and imputed data

Description

Scatterplot of observed and imputed data

Usage

xyplot(...)

Arguments

...	Any arguments passed to the function.

Value

The output of mice::xyplot and a message about the ggmice equivalent.

Examples

imp <- mice::mice(mice::nhanes, maxit = 1, printFlag = FALSE)
xyplot(imp, bmi ~ age)

---