Function reference • naniar

Data structures for missing data

Creation and Manipulation of Shadow Matrices

as_shadow(): Create shadows

as_shadow_upset(): Convert data into shadow format for doing an upset plot

bind_shadow(): Bind a shadow dataframe to original data

nabular(): Convert data into nabular form by binding shade to it

gather_shadow(): Long form representation of a shadow matrix

shade(): Create new levels of missing

shadow_long(): Reshape shadow data into a long format

unbind_shadow() unbind_data(): Unbind (remove) shadow from data, and vice versa

shadow_shift(): Shift missing values to facilitate missing data exploration/visualisation

Create special missing values

Create special missing values so that they don’t get lost! See vignette("special-missing").

recode_shadow(): Add special missing values to the shadow matrix

Visualisation

Visualise missing data

geom_miss_point(): Plot Missing Data Points

stat_miss_point(): stat_miss_point

gg_miss_case(): Plot the number of missings per case (row)

gg_miss_case_cumsum(): Plot of cumulative sum of missing for cases

gg_miss_fct(): Plot the number of missings for each variable, broken down by a factor

gg_miss_span(): Plot the number of missings in a given repeating span

gg_miss_upset(): Plot the pattern of missingness using an upset plot.

gg_miss_var(): Plot the number of missings for each variable

gg_miss_var_cumsum(): Plot of cumulative sum of missing value for each variable

gg_miss_which(): Plot which variables contain a missing value

reexports %>% is_na are_na vis_miss: Objects exported from other packages

Numerical Summaries

Provide tidy data frame summaries of missingness

miss_var_prop() complete_var_prop() miss_var_pct() complete_var_pct() miss_case_prop() complete_case_prop() miss_case_pct() complete_case_pct(): Proportion of variables containing missings or complete values

miss_case_cumsum(): Summarise the missingness in each case

miss_case_summary(): Summarise the missingness in each case

miss_case_table(): Tabulate missings in cases.

miss_prop_summary(): Proportions of missings in data, variables, and cases.

miss_scan_count(): Search and present different kinds of missing values

miss_summary(): Collate summary measures from naniar into one tibble

miss_var_cumsum(): Cumulative sum of the number of missings in each variable

miss_var_run(): Find the number of missing and complete values in a single run

miss_var_span(): Summarise the number of missings for a given repeating span on a variable

miss_var_summary(): Summarise the missingness in each variable

miss_var_table(): Tabulate the missings in the variables

miss_var_which(): Which variables contain missing values?

Handy helpers

n_var_complete() n_case_complete(): The number of variables with complete values

n_var_miss() n_case_miss(): The number of variables or cases with missing values

n_complete(): Return the number of complete values

n_complete_row(): Return a vector of the number of complete values in each row

n_miss(): Return the number of missing values

n_miss_row(): Return a vector of the number of missing values in each row

prop_miss_case() prop_complete_case(): Proportion of cases that contain a missing or complete values.

prop_miss_var() prop_complete_var(): Proportion of variables containing missings or complete values

prop_complete(): Return the proportion of complete values

prop_complete_row(): Return a vector of the proportion of missing values in each row

prop_miss(): Return the proportion of missing values

prop_miss_row(): Return a vector of the proportion of missing values in each row

pct_miss_case() pct_complete_case(): Percentage of cases that contain a missing or complete values.

pct_miss_var() pct_complete_var(): Percentage of variables containing missings or complete values

pct_complete(): Return the percent of complete values

pct_miss(): Return the percent of missing values

any_na() any_miss() any_complete() all_na() all_miss() all_complete(): Identify if there are any or all missing or complete values

any_row_miss(): Helper function to determine whether there are any missings

is_shade() are_shade() any_shade(): Detect if this is a shade

which_are_shade(): Which variables are shades?

common_na_numbers: Common number values for NA

common_na_strings: Common string values for NA

Add columns

Add missing data summaries/tool columns

add_any_miss(): Add a column describing presence of any missing values

add_label_missings(): Add a column describing if there are any missings in the dataset

add_label_shadow(): Add a column describing whether there is a shadow

add_miss_cluster(): Add a column that tells us which "missingness cluster" a row belongs to

add_n_miss(): Add column containing number of missing data values

add_prop_miss(): Add column containing proportion of missing data values

add_shadow(): Add a shadow column to dataframe

add_shadow_shift(): Add a shadow shifted column to a dataset

add_span_counter(): Add a counter variable for a span of dataframe

Replacing values with and to NA

Functions to help replace certain values with NA, which includes scoped variants (_at, _if, _all) that take formulas for flexible approachs. vignette("replace-with-na")

replace_with_na(): Replace values with missings

replace_with_na_all(): Replace all values with NA where a certain condition is met

replace_with_na_at(): Replace specified variables with NA where a certain condition is met

replace_with_na_if(): Replace values with NA based on some condition, for variables that meet some predicate

replace_to_na(): Replace values with missings

replace_na_with(): Replace NA value with provided value

Imputation helpers

Simple imputation methods for exploring visualisation and missingness structure. See vignette("exploring-imputed-values") for more details.

impute_below(): Impute data with values shifted 10 percent below range.

impute_below(<numeric>): Impute numeric values below a range for graphical exploration

impute_below_all(): Impute data with values shifted 10 percent below range.

impute_below_at(): Scoped variants of impute_below

impute_below_if(): Scoped variants of impute_below

impute_factor(): Impute a factor value into a vector with missing values

impute_fixed(): Impute a fixed value into a vector with missing values

impute_mean(): Impute the mean value into a vector with missing values

impute_median(): Impute the median value into a vector with missing values

impute_mode(): Impute the mode value into a vector with missing values

impute_zero(): Impute zero into a vector with missing values

impute_mean_all() impute_mean_at() impute_mean_if(): Scoped variants of impute_mean

impute_median_all() impute_median_at() impute_median_if(): Scoped variants of impute_median

set_prop_miss() set_n_miss(): Set a proportion or number of missing values

Package title details

Details of the package naniar

naniar-package naniar: naniar: Data Structures, Summaries, and Visualisations for Missing Data

Cast Shadows

Add shadow information to the dataframe while reducing it to the variables of interest

cast_shadow(): Add a shadow column to a dataset

cast_shadow_shift(): Add a shadow and a shadow_shift column to a dataset

cast_shadow_shift_label(): Add a shadow column and a shadow shifted column to a dataset

Misc helpers

label_miss_1d(): Label a missing from one column

label_miss_2d(): label_miss_2d

label_missings(): Is there a missing value in the row of a dataframe?

where_na(): Which rows and cols contain missings?

which_na(): Which elements contain missings?

.where(): Split a call into two components with a useful verb name

Data Sources

For practice and example usecases in naniar

oceanbuoys: West Pacific Tropical Atmosphere Ocean Data, 1993 & 1997.

pedestrian: Pedestrian count information around Melbourne for 2016

riskfactors: The Behavioral Risk Factor Surveillance System (BRFSS) Survey Data, 2009.

Little’s MCAR test

For performing Little’s MCAR test

mcar_test(): Little's missing completely at random (MCAR) test

ggplot2 extensions

Custom ggplot geoms built to extend ggplot for missing values

StatMissPoint: naniar-ggproto