Imputation of missing values
imputes the missing value with imputate_na()
imputate_na() imputes the missing value contained in the variable. The predictor with missing values support both numeric and categorical variables, and supports the following method.
- predictor is numerical variable
- “mean” : arithmetic mean
- “median” : median
- “mode” : mode
- “knn” : K-nearest neighbors
- target variable must be specified
- “rpart” : Recursive Partitioning and Regression Trees
- target variable must be specified
- target variable must be specified
- “mice” : Multivariate Imputation by Chained Equations
- target variable must be specified
- random seed must be set
- target variable must be specified
- predictor is categorical variable
- “mode” : mode
- “rpart” : Recursive Partitioning and Regression Trees
- target variable must be specified
- target variable must be specified
- “mice” : Multivariate Imputation by Chained Equations
- target variable must be specified
- random seed must be set
- target variable must be specified
In the following example, imputate_na() imputes the missing value of Income, a numeric variable of carseats, using the “rpart” method. summary() summarizes missing value imputation information, and plot() visualizes missing information.
if (requireNamespace("rpart", quietly = TRUE)) {
income <- imputate_na(carseats, Income, US, method = "rpart")
# result of imputation
income
# summary of imputation
summary(income)
# viz of imputation
plot(income)
} else {
cat("If you want to use this feature, you need to install the rpart package.\n")
}
* Impute missing values based on Recursive Partitioning and Regression Trees
- method : rpart
* Information of Imputation (before vs after)
Original Imputation
described_variables "value" "value"
n "380" "400"
na "20" " 0"
mean "68.86053" "69.05073"
sd "28.09161" "27.57382"
se_mean "1.441069" "1.378691"
IQR "48.25" "46.00"
skewness "0.04490600" "0.02935732"
kurtosis "-1.089201" "-1.035086"
p00 "21" "21"
p01 "21.79" "21.99"
p05 "26" "26"
p10 "30.0" "30.9"
p20 "39" "40"
p25 "42.75" "44.00"
p30 "48.00000" "51.58333"
p40 "62" "63"
p50 "69" "69"
p60 "78.0" "77.4"
p70 "86.3" "84.3"
p75 "91" "90"
p80 "96.2" "96.0"
p90 "108.1" "106.1"
p95 "115.05" "115.00"
p99 "119.21" "119.01"
p100 "120" "120" 
The following imputes the categorical variable urban by the “mice” method.
library(mice)
다음의 패키지를 부착합니다: 'mice'
The following object is masked from 'package:stats':
filter
The following objects are masked from 'package:base':
cbind, rbind
urban <- imputate_na(carseats, Urban, US, method = "mice")
iter imp variable
1 1 Income Urban
1 2 Income Urban
1 3 Income Urban
1 4 Income Urban
1 5 Income Urban
2 1 Income Urban
2 2 Income Urban
2 3 Income Urban
2 4 Income Urban
2 5 Income Urban
3 1 Income Urban
3 2 Income Urban
3 3 Income Urban
3 4 Income Urban
3 5 Income Urban
4 1 Income Urban
4 2 Income Urban
4 3 Income Urban
4 4 Income Urban
4 5 Income Urban
5 1 Income Urban
5 2 Income Urban
5 3 Income Urban
5 4 Income Urban
5 5 Income Urban
# result of imputation
urban
[1] Yes Yes Yes Yes Yes No Yes Yes No No No Yes Yes Yes Yes No Yes Yes
[19] No Yes Yes No Yes Yes Yes No No Yes Yes Yes Yes Yes Yes Yes Yes No
[37] No Yes Yes No No Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes
[55] No Yes Yes Yes Yes Yes Yes No Yes Yes No No Yes Yes Yes Yes Yes No
[73] Yes No No No Yes No Yes Yes Yes Yes Yes No No No Yes No Yes No
[91] No Yes Yes No Yes Yes No Yes No No No Yes No Yes Yes Yes No Yes
[109] Yes No Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes No
[127] Yes Yes Yes No Yes Yes Yes Yes Yes No No Yes Yes No Yes Yes Yes Yes
[145] No Yes Yes No No Yes No No No No No Yes Yes No No No No No
[163] Yes No No Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes No Yes No Yes
[181] Yes Yes Yes Yes No Yes No Yes Yes No No Yes No Yes Yes Yes Yes Yes
[199] Yes Yes No Yes No Yes Yes Yes Yes No Yes No No Yes Yes Yes Yes Yes
[217] Yes No Yes Yes Yes Yes Yes Yes No Yes Yes Yes No No No No Yes No
[235] No Yes Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Yes Yes Yes Yes Yes
[253] Yes No Yes Yes Yes Yes No No Yes Yes Yes Yes Yes Yes No No Yes Yes
[271] Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes No Yes No No Yes No Yes
[289] No Yes No No Yes Yes Yes No Yes Yes Yes No Yes Yes Yes Yes Yes Yes
[307] Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes No No No Yes Yes Yes Yes
[325] Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes No Yes Yes No
[343] No Yes No Yes No No Yes No No No Yes No Yes Yes Yes Yes Yes Yes
[361] No No Yes Yes Yes No No Yes No Yes Yes Yes No Yes Yes Yes Yes No
[379] Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes No Yes Yes
[397] No Yes Yes Yes
attr(,"var_type")
[1] categorical
attr(,"method")
[1] mice
attr(,"na_pos")
[1] 33 36 84 94 113 132 151 292 313 339
attr(,"seed")
[1] 24283
attr(,"type")
[1] missing values
attr(,"message")
[1] complete imputation
attr(,"success")
[1] TRUE
Levels: No Yes
# summary of imputation
summary(urban)
* Impute missing values based on Multivariate Imputation by Chained Equations
- method : mice
- random seed : 24283
* Information of Imputation (before vs after)
original imputation original_percent imputation_percent
No 115 122 28.75 30.5
Yes 275 278 68.75 69.5
<NA> 10 0 2.50 0.0
# viz of imputation
plot(urban)
Collaboration with dplyr
The following example imputes the missing value of the Income variable, and then calculates the arithmetic mean for each level of US. In this case, dplyr is used, and it is easily interpreted logically using pipes.
# The mean before and after the imputation of the Income variable
carseats %>%
mutate(Income_imp = imputate_na(carseats, Income, US, method = "knn")) %>%
group_by(US) %>%
summarise(orig = mean(Income, na.rm = TRUE),
imputation = mean(Income_imp))
# A tibble: 2 × 3
US orig imputation
<fct> <dbl> <dbl>
1 No 65.8 66.1
2 Yes 70.4 70.5
