| Title: | Missing Data Explorer |
|---|---|
| Description: | Correct identification and handling of missing data is one of the most important steps in any analysis. To aid this process, 'mde' provides a very easy to use yet robust framework to quickly get an idea of where the missing data lies and therefore find the most appropriate action to take. Graham WJ (2009) <doi:10.1146/annurev.psych.58.110405.085530>. |
| Authors: | Nelson Gonzabato [aut, cre] |
| Maintainer: | Nelson Gonzabato <[email protected]> |
| License: | GPL-3 |
| Version: | 0.3.2 |
| Built: | 2024-11-12 03:40:22 UTC |
| Source: | https://github.com/nelson-gon/mde |
This is a helper function to check if all column/vector values are NA
all_na(x)all_na(x)
x |
A vector or data.frame column |
Boolean TRUE or FALSE depending on the nature of the column/vector
test <- data.frame(A=c(NA, 2), B= c(NA, NA)) all_na(test) test_vec <- c("NA",NA,"nope") test_numeric <- c(NA, 2) all_na(test_vec) all_na(test_numeric)test <- data.frame(A=c(NA, 2), B= c(NA, NA)) all_na(test) test_vec <- c("NA",NA,"nope") test_numeric <- c(NA, 2) all_na(test_vec) all_na(test_numeric)
Recode NA as based on Other Columns
column_based_recode( df, criterion = "all_na", values_from = NULL, values_to = NULL, value = 0, pattern_type = "contains", pattern = "Solar", case_sensitive = FALSE )column_based_recode( df, criterion = "all_na", values_from = NULL, values_to = NULL, value = 0, pattern_type = "contains", pattern = "Solar", case_sensitive = FALSE )
df |
A data.frame object for which recoding is to be done. |
criterion |
Currently supports one of all_na or any_na to index rows that are either all NA or contain any NA. |
values_from |
Character. Name of column to get the original values from |
values_to |
Character New column name for the newly recoded values. Defaults to the same name if none is supplied. |
value |
The value to convert to 'NA'. We can for instance change "n/a" to 'NA' or any other value. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
A 'data.frame' object with target 'NA' values replaced.
df <- structure(list(id = 40:43, v1 = c(NA, 1L, 1L, 1L), v2 = c(NA, 1L, 1L, 1L), v3 = c(NA, 2L, NA, 1L), test = c(1L, 2L, 1L, 3L)), class = "data.frame", row.names = c(NA, -4L)) # recode test as 0 if all NA, return test otherwise column_based_recode(df,values_from = "test", pattern_type = "starts_with", pattern="v")df <- structure(list(id = 40:43, v1 = c(NA, 1L, 1L, 1L), v2 = c(NA, 1L, 1L, 1L), v3 = c(NA, 2L, NA, 1L), test = c(1L, 2L, 1L, 3L)), class = "data.frame", row.names = c(NA, -4L)) # recode test as 0 if all NA, return test otherwise column_based_recode(df,values_from = "test", pattern_type = "starts_with", pattern="v")
Recode NA as another value using a function or a custom equation
custom_na_recode( df, func = "mean", grouping_cols = NULL, across_columns = NULL )custom_na_recode( df, func = "mean", grouping_cols = NULL, across_columns = NULL )
df |
A valid R 'object' for which the percentage of missing values is required. |
func |
Function to use for the replacement e.g "mean". Defaults to mean. |
grouping_cols |
A character vector. If supplied, one can provide the columns by which to group the data. |
across_columns |
A character vector specifying across which columns recoding should be done #use all columns head(custom_na_recode(airquality,func="mean")) # use only a few columns head(custom_na_recode(airquality,func="mean",across_columns = c("Solar.R","Ozone"))) # use a function from another package #head(custom_na_recode(airquality, func=dplyr::lead)) some_data <- data.frame(ID=c("A1","A1","A1","A2","A2", "A2"), A=c(5,NA,0,8,3,4), B=c(10,0,0,NA,5,6),C=c(1,NA,NA,25,7,8)) # grouping head(custom_na_recode(some_data,func = "mean", grouping_cols = "ID", across_columns = c("C", "A"))) head(custom_na_recode(some_data,func = "mean", grouping_cols = "ID")) |
Recode Missing Values Dictionary-Style
dict_recode( df, use_func = "recode_na_as", pattern_type = "starts_with", patterns, values )dict_recode( df, use_func = "recode_na_as", pattern_type = "starts_with", patterns, values )
df |
A data.frame object for which recoding is to be done. |
use_func |
Function to use for the recoding. One of the various 'recode_*' functions in package 'mde'. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
patterns |
A vector containing patterns to use for pattern_type |
values |
A vector containing values to match to the patterns vector |
A 'data.frame' object with replacements as required.
head(dict_recode(airquality, pattern_type="starts_with", patterns = c("Solar", "Ozone"), values = c(190, 41), use_func="recode_as_na")) head(dict_recode(airquality, pattern_type="starts_with", patterns = c("Solar", "Ozone"), values = c(42, 420), use_func="recode_na_as"))head(dict_recode(airquality, pattern_type="starts_with", patterns = c("Solar", "Ozone"), values = c(190, 41), use_func="recode_as_na")) head(dict_recode(airquality, pattern_type="starts_with", patterns = c("Solar", "Ozone"), values = c(42, 420), use_func="recode_na_as"))
Drop columns for which all values are NA
drop_all_na(df, grouping_cols = NULL)drop_all_na(df, grouping_cols = NULL)
df |
A valid R 'object' for which the percentage of missing values is required. |
grouping_cols |
A character vector. If supplied, one can provide the columns by which to group the data. |
test <- data.frame(ID= c("A","A","B","A","B"), Vals = c(rep(NA,4),2)) test2 <- data.frame(ID= c("A","A","B","A","B"), Vals = rep(NA, 5)) # drop columns where all values are NA drop_all_na(test2) # drop NAs only if all are NA for a given group, drops group too. drop_all_na(test, "ID")test <- data.frame(ID= c("A","A","B","A","B"), Vals = c(rep(NA,4),2)) test2 <- data.frame(ID= c("A","A","B","A","B"), Vals = rep(NA, 5)) # drop columns where all values are NA drop_all_na(test2) # drop NAs only if all are NA for a given group, drops group too. drop_all_na(test, "ID")
Provides a simple yet efficient way to drop missing values("NA"s) at columns that match a given pattern.
drop_na_at( df, pattern_type = "contains", pattern = NULL, case_sensitive = FALSE, ... )drop_na_at( df, pattern_type = "contains", pattern = NULL, case_sensitive = FALSE, ... )
df |
A data.frame object |
pattern_type |
One of "contains", "ends_with" or "starts_with" |
pattern |
The type of pattern to use when matching the pattern_type. The pattern is case sensitive |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other params to other methods |
A data.frame object containing only columns that match the given pattern with the missing values removed.
head(drop_na_at(airquality,pattern_type = "starts_with","O"))head(drop_na_at(airquality,pattern_type = "starts_with","O"))
"drop_na_if" provides a simple way to drop columns with missing values if they meet certain criteria/conditions.
drop_na_if( df, sign = "gteq", percent_na = 50, keep_columns = NULL, grouping_cols = NULL, target_columns = NULL, ... )drop_na_if( df, sign = "gteq", percent_na = 50, keep_columns = NULL, grouping_cols = NULL, target_columns = NULL, ... )
df |
A data.frame object |
sign |
Character. One of gteq,lteq,lt,gt or eq which refer to greater than(gt) or equal(eq) or less than(lt) or equal to(eq) respectively. |
percent_na |
The percentage to use when dropping columns with missing values |
keep_columns |
Columns that should be kept despite meeting the target percent_na criterion(criteria) |
grouping_cols |
For dropping groups that meet a target criterion of percent missingness. |
target_columns |
If working on grouped data, drop all columns that meet target or only a specific column. |
... |
Other arguments to "percent_missing" |
A data.frame object with columns that meet the target criteria dropped.
head(drop_na_if(airquality, percent_na = 24)) #drop columns that have less tan or equal to 4% head(drop_na_if(airquality,sign="lteq", percent_na = 4)) # Drop all except with greater than ie equal to 4% missing but keep Ozone head(drop_na_if(airquality, sign="gteq",percent_na = 4, keep_columns = "Ozone")) # Drop groups that meet a given criterion grouped_drop <- structure(list(ID = c("A", "A", "B", "A", "B"), Vals = c(4, NA, NA, NA, NA), Values = c(5, 6, 7, 8, NA)), row.names = c(NA, -5L), class = "data.frame") drop_na_if(grouped_drop,percent_na = 67,grouping_cols = "ID")head(drop_na_if(airquality, percent_na = 24)) #drop columns that have less tan or equal to 4% head(drop_na_if(airquality,sign="lteq", percent_na = 4)) # Drop all except with greater than ie equal to 4% missing but keep Ozone head(drop_na_if(airquality, sign="gteq",percent_na = 4, keep_columns = "Ozone")) # Drop groups that meet a given criterion grouped_drop <- structure(list(ID = c("A", "A", "B", "A", "B"), Vals = c(4, NA, NA, NA, NA), Values = c(5, 6, 7, 8, NA)), row.names = c(NA, -5L), class = "data.frame") drop_na_if(grouped_drop,percent_na = 67,grouping_cols = "ID")
Conditionally drop rows based on percent missingness
drop_row_if(df, sign = "gt", type = "count", value = 20, as_percent = TRUE)drop_row_if(df, sign = "gt", type = "count", value = 20, as_percent = TRUE)
df |
A data.frame object |
sign |
Character. One of gteq,lteq,lt,gt or eq which refer to greater than(gt) or equal(eq) or less than(lt) or equal to(eq) respectively. |
type |
One of either count or percent. Defaults to count |
value |
Value to use for the drop. |
as_percent |
Logical. If set to TRUE, percent_na is treated as a percentage. Otherwise, decimals(fractions) are used. |
head(drop_row_if(airquality,sign = "gteq", type = "percent",value=16, as_percent = TRUE)) # should give the same output as above. head(drop_row_if(airquality, sign="gteq", type="percent",value = 0.15, as_percent=FALSE)) # Drop based on NA counts df <- data.frame(A=1:5, B=c(1,NA,NA,2, 3), C= c(1,NA,NA,2,3)) drop_row_if(df, type="count",value=2,sign="eq")head(drop_row_if(airquality,sign = "gteq", type = "percent",value=16, as_percent = TRUE)) # should give the same output as above. head(drop_row_if(airquality, sign="gteq", type="percent",value = 0.15, as_percent=FALSE)) # Drop based on NA counts df <- data.frame(A=1:5, B=c(1,NA,NA,2, 3), C= c(1,NA,NA,2,3)) drop_row_if(df, type="count",value=2,sign="eq")
This function takes a 'data.frame' object as an input and returns the corresponding ‘NA' counts. 'NA' refers to R’s builtin missing data holder.
get_na_counts(x, grouping_cols = NULL, exclude_cols = NULL)get_na_counts(x, grouping_cols = NULL, exclude_cols = NULL)
x |
A valid R 'object' for which 'na_counts' are needed. |
grouping_cols |
A character vector. If supplied, one can provide the columns by which to group the data. |
exclude_cols |
Columns to exclude from the analysis. |
An object of the same type as 'x' showing the respective number of missing values. If grouped is set to 'TRUE', the results are returned by group.
get_na_counts(airquality) # Grouped counts test <- data.frame(Subject = c("A","A","B","B"), res = c(NA,1,2,3), ID = c("1","1","2","2")) get_na_counts(test,grouping_cols = c("ID", "Subject"))get_na_counts(airquality) # Grouped counts test <- data.frame(Subject = c("A","A","B","B"), res = c(NA,1,2,3), ID = c("1","1","2","2")) get_na_counts(test,grouping_cols = c("ID", "Subject"))
Get mean missingness.
get_na_means(x, as_percent = TRUE)get_na_means(x, as_percent = TRUE)
x |
A vector whose mean NA is required. |
as_percent |
Boolean? Report means as percents, defaults to TRUE. |
get_na_means(airquality)get_na_means(airquality)
Get NA counts for a given character, numeric, factor, etc.
na_counts(x)na_counts(x)
x |
A vector whose number of missing values is to be determined. |
na_counts(airquality$Ozone)na_counts(airquality$Ozone)
An all-in-one missingness report
na_summary( df, grouping_cols = NULL, sort_by = NULL, descending = FALSE, exclude_cols = NULL, pattern = NULL, pattern_type = NULL, regex_kind = "exclusion", round_to = NULL, reset_rownames = FALSE )na_summary( df, grouping_cols = NULL, sort_by = NULL, descending = FALSE, exclude_cols = NULL, pattern = NULL, pattern_type = NULL, regex_kind = "exclusion", round_to = NULL, reset_rownames = FALSE )
df |
A valid R 'object' for which the percentage of missing values is required. |
grouping_cols |
A character vector. If supplied, one can provide the columns by which to group the data. |
sort_by |
One of counts or percents. This determines whether the results are sorted by counts or percentages. |
descending |
Logical. Should missing values be sorted in decreasing order ie largest to smallest? Defaults to FALSE. |
exclude_cols |
A character vector indicating columns to exclude when returning results. |
pattern |
Pattern to use for exclusion or inclusion. column inclusion criteria. |
pattern_type |
A regular expression type. One of "starts_with", "contains", or "regex". Defaults to NULL. Only use for selective inclusion. |
regex_kind |
One of inclusion or exclusion. Defaults to exclusion to exclude columns using regular expressions. |
round_to |
Number of places to round 2. Defaults to user digits option. |
reset_rownames |
Should the rownames be reset in the output? defaults to FALSE |
na_summary(airquality) # grouping test2 <- data.frame(ID= c("A","A","B","A","B"),Vals = c(rep(NA,4),"No"), ID2 = c("E","E","D","E","D")) df <- data.frame(A=1:5,B=c(NA,NA,25,24,53), C=c(NA,1,2,3,4)) na_summary(test2,grouping_cols = c("ID","ID2")) # sort summary na_summary(airquality,sort_by = "percent_missing",descending = TRUE) na_summary(airquality,sort_by = "percent_complete") # Include only via a regular expression na_summary(mtcars, pattern_type = "contains", pattern = "mpg|disp|wt", regex_kind = "inclusion") na_summary(airquality, pattern_type = "starts_with", pattern = "ozone", regex_kind = "inclusion") # exclusion via a regex na_summary(airquality, pattern_type = "starts_with", pattern = "oz|Sol", regex_kind = "exclusion") # reset rownames when sorting by variable na_summary(df,sort_by="variable",descending=TRUE, reset_rownames = TRUE)na_summary(airquality) # grouping test2 <- data.frame(ID= c("A","A","B","A","B"),Vals = c(rep(NA,4),"No"), ID2 = c("E","E","D","E","D")) df <- data.frame(A=1:5,B=c(NA,NA,25,24,53), C=c(NA,1,2,3,4)) na_summary(test2,grouping_cols = c("ID","ID2")) # sort summary na_summary(airquality,sort_by = "percent_missing",descending = TRUE) na_summary(airquality,sort_by = "percent_complete") # Include only via a regular expression na_summary(mtcars, pattern_type = "contains", pattern = "mpg|disp|wt", regex_kind = "inclusion") na_summary(airquality, pattern_type = "starts_with", pattern = "ozone", regex_kind = "inclusion") # exclusion via a regex na_summary(airquality, pattern_type = "starts_with", pattern = "oz|Sol", regex_kind = "exclusion") # reset rownames when sorting by variable na_summary(df,sort_by="variable",descending=TRUE, reset_rownames = TRUE)
A convenient way to obtain percent missingness column-wise.
percent_missing(df, grouping_cols = NULL, exclude_cols = NULL)percent_missing(df, grouping_cols = NULL, exclude_cols = NULL)
df |
A valid R 'object' for which the percentage of missing values is required. |
grouping_cols |
A character vector. If supplied, one can provide the columns by which to group the data. |
exclude_cols |
A character vector indicating columns to exclude when returning results. |
An object of the same class as x showing the percentage of missing values.
test <- data.frame(ID= c("A","B","A","B","A","B","A"), Vals = c(NA,25,34,NA,67,NA,45)) percent_missing(test,grouping_cols = "ID") percent_missing(airquality) percent_missing(airquality,exclude_cols = c("Day","Temp"))test <- data.frame(ID= c("A","B","A","B","A","B","A"), Vals = c(NA,25,34,NA,67,NA,45)) percent_missing(test,grouping_cols = "ID") percent_missing(airquality) percent_missing(airquality,exclude_cols = c("Day","Temp"))
percent missing but for vectors.
percent_na(x)percent_na(x)
x |
A vector whose mean NA is required. |
percent_na(airquality$Ozone)percent_na(airquality$Ozone)
This provides a convenient way to convert a number/value that should indeed be an "NA" to "NA". In otherwords, it converts a value to R's recognized NA.
recode_as_na( df, value = NULL, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )recode_as_na( df, value = NULL, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )
df |
A data.frame object for which recoding is to be done. |
value |
The value to convert to 'NA'. We can for instance change "n/a" to 'NA' or any other value. |
subset_cols |
An optional character vector to define columns for which changes are required. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to other functions |
An object of the same class as x with values changed to 'NA'.
head(recode_as_na(airquality,value=c(67,118),pattern_type="starts_with",pattern="S|O")) head(recode_as_na(airquality,value=c(41),pattern_type="ends_with",pattern="e")) head(recode_as_na(airquality, value=41,subset_cols="Ozone"))head(recode_as_na(airquality,value=c(67,118),pattern_type="starts_with",pattern="S|O")) head(recode_as_na(airquality,value=c(41),pattern_type="ends_with",pattern="e")) head(recode_as_na(airquality, value=41,subset_cols="Ozone"))
Recode Values as NA if they meet defined criteria
recode_as_na_for(df, criteria = "gt", value = 0, subset_cols = NULL)recode_as_na_for(df, criteria = "gt", value = 0, subset_cols = NULL)
df |
A data.frame object to manipulate |
criteria |
One of gt,gteq,lt,lteq to define greater than, greater than or equal to, less than or less than or equal to. |
value |
The value to convert to 'NA'. We can for instance change "n/a" to 'NA' or any other value. |
subset_cols |
An optional character vector for columns to manipulate. |
A data.frame object with the required changes.
recode_as_na_for(airquality,value=36, criteria = "gteq", subset_cols = c("Ozone","Solar.R"))recode_as_na_for(airquality,value=36, criteria = "gteq", subset_cols = c("Ozone","Solar.R"))
Conditionally change all column values to NA
recode_as_na_if(df, sign = "gteq", percent_na = 50, keep_columns = NULL, ...)recode_as_na_if(df, sign = "gteq", percent_na = 50, keep_columns = NULL, ...)
df |
A data.frame object |
sign |
Character. One of gteq,lteq,lt,gt or eq which refer to greater than(gt) or equal(eq) or less than(lt) or equal to(eq) respectively. |
percent_na |
The percentage to use when dropping columns with missing values |
keep_columns |
Columns that should be kept despite meeting the target percent_na criterion(criteria) |
... |
Other arguments to "percent_missing" |
A 'data.frame' with the target columns populated with 'NA's.
head(recode_as_na_if(airquality, sign="gt", percent_na=20))head(recode_as_na_if(airquality, sign="gt", percent_na=20))
Recode as NA based on string match
recode_as_na_str( df, pattern_type = "ends_with", pattern = NULL, case_sensitive = FALSE, ... )recode_as_na_str( df, pattern_type = "ends_with", pattern = NULL, case_sensitive = FALSE, ... )
df |
A data.frame object |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to grepl |
partial_match <- data.frame(A=c("Hi","match_me","nope"), B=c(NA, "not_me","nah")) # Replace all that end with "me" with NA recode_as_na_str(partial_match,"ends_with","me") # Do not recode, ie case-sensitive recode_as_na_str(partial_match,"ends_with","ME", case_sensitive=TRUE)partial_match <- data.frame(A=c("Hi","match_me","nope"), B=c(NA, "not_me","nah")) # Replace all that end with "me" with NA recode_as_na_str(partial_match,"ends_with","me") # Do not recode, ie case-sensitive recode_as_na_str(partial_match,"ends_with","ME", case_sensitive=TRUE)
This provides a convenient way to convert a number/value to another value.
recode_as_value( df, value = NULL, replacement_value = NULL, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )recode_as_value( df, value = NULL, replacement_value = NULL, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )
df |
A data.frame object for which recoding is to be done. |
value |
The value/vector of values to convert. |
replacement_value |
New value. |
subset_cols |
An optional character vector to define columns for which changes are required. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to other functions |
An object of the same class as x with values changed to 'NA'.
head(recode_as_value(airquality, value=c(67,118),replacement=NA, pattern_type="starts_with",pattern="S|O"))head(recode_as_value(airquality, value=c(67,118),replacement=NA, pattern_type="starts_with",pattern="S|O"))
Helper functions in package mde
recode_helper( x, pattern_type = NULL, pattern = NULL, original_value, new_value, case_sensitive = FALSE, ... )recode_helper( x, pattern_type = NULL, pattern = NULL, original_value, new_value, case_sensitive = FALSE, ... )
x |
A data.frame object |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
original_value |
Value to replace |
new_value |
Replacement value. |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to other functions |
This provides a convenient way to recode "NA" as another value for instance "NaN", "n/a" or any other value a user wishes to use.
recode_na_as( df, value = 0, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )recode_na_as( df, value = 0, subset_cols = NULL, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )
df |
A data.frame object for which recoding is to be done. |
value |
The value to convert to 'NA'. We can for instance change "n/a" to 'NA' or any other value. |
subset_cols |
An optional character vector to define columns for which changes are required. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to other functions |
An object of the same type as x with NAs replaced with the desired value.
head(recode_na_as(airquality, "n/a")) head(recode_na_as(airquality, subset_cols = "Ozone", value = "N/A")) head(recode_na_as(airquality, value=0, pattern_type="starts_with", pattern="Solar"))head(recode_na_as(airquality, "n/a")) head(recode_na_as(airquality, subset_cols = "Ozone", value = "N/A")) head(recode_na_as(airquality, value=0, pattern_type="starts_with", pattern="Solar"))
Recode NA as another value with some conditions
recode_na_if(df, grouping_cols = NULL, target_groups = NULL, replacement = 0)recode_na_if(df, grouping_cols = NULL, target_groups = NULL, replacement = 0)
df |
A data.frame object with missing values |
grouping_cols |
Character columns to use for grouping the data |
target_groups |
Character Recode NA as if and only if the grouping column is in this vector of values |
replacement |
Values to use to replace NAs for IDs that meet the requirements. Defaults to 0. |
some_data <- data.frame(ID=c("A1","A2","A3", "A4"), A=c(5,NA,0,8), B=c(10,0,0,1),C=c(1,NA,NA,25)) # Replace NAs with 0s only for IDs in A2 and A3 recode_na_if(some_data,"ID",c("A2","A3"),replacement=0)some_data <- data.frame(ID=c("A1","A2","A3", "A4"), A=c(5,NA,0,8), B=c(10,0,0,1),C=c(1,NA,NA,25)) # Replace NAs with 0s only for IDs in A2 and A3 recode_na_if(some_data,"ID",c("A2","A3"),replacement=0)
Helper functions in package mde
recode_selectors( x, column_check = TRUE, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )recode_selectors( x, column_check = TRUE, pattern_type = NULL, pattern = NULL, case_sensitive = FALSE, ... )
x |
data.frame object |
column_check |
If TRUE, pattern search is performed columnwise. Defaults to FALSE. |
pattern_type |
One of contains', 'starts_with' or 'ends_with'. |
pattern |
A character pattern to match |
case_sensitive |
Defaults to FALSE. Patterns are case insensitive if TRUE |
... |
Other arguments to other functions |
Provides a useful way to sort the variables(columns) according to their missingness.
sort_by_missingness(df, sort_by = "counts", descending = FALSE, ...)sort_by_missingness(df, sort_by = "counts", descending = FALSE, ...)
df |
A data.frame object |
sort_by |
One of counts or percents. This determines whether the results are sorted by counts or percentages. |
descending |
Logical. Should missing values be sorted in decreasing order ie largest to smallest? Defaults to FALSE. |
... |
Other arguments to specific functions. See "See also below" |
A 'data.frame' object sorted by number/percentage of missing values
sort_by_missingness(airquality, sort_by = "counts") # sort by percents sort_by_missingness(airquality, sort_by="percents") # descending order sort_by_missingness(airquality, descend = TRUE)sort_by_missingness(airquality, sort_by = "counts") # sort by percents sort_by_missingness(airquality, sort_by="percents") # descending order sort_by_missingness(airquality, descend = TRUE)