imputeTestbench: Test Bench for Missing Data Imputing Models/Methods Comparison
> library(imputeTestbench)
バージョン: 1.1.0
| 関数名 | 概略 |
|---|---|
append_method |
To attach and compare new method to existing comparison study done with function 'impute_errors()' |
impute_errors |
Function working as testbench for comparison of imputing models |
mae |
Mean Absolute Error Calculation |
mape |
Mean Absolute Percent Error Calculation |
plot_errors |
Function to plot the Error Comparison |
remove_method |
Removes unwanted method from already existing methods in Test bench |
rmse |
Root Mean Square Error Calculation |
impute_errors
> impute_errors()
$Parameter
[1] "RMSE Plot"
$Missing_Percent
[1] 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
$Historic_Mean
[1] 0.4908711 0.6641892 0.8008869 0.9175113 1.0210202 1.1151930 1.2025538
[8] 1.2857627
$Interpolation
[1] 0.5567764 0.7810250 0.9539392 1.1000000 1.2288206 1.3453624 1.4525839
[8] 1.5524175
mae
> x <- sample(1:7, 100, replace = TRUE)
> y <- sample(1:4, 100, replace = TRUE)
> mae(x - y)
[1] 2.13
mape
> x <- sample(1:7, 100, replace = TRUE)
> y <- sample(1:4, 100, replace = TRUE)
> mape((x - y), x)
[1] 55.70476
plot_errors
> impute_errors() %>% plot_errors()
rmse
> x <- sample(1:7, 100, replace = TRUE)
> y <- sample(1:4, 100, replace = TRUE)
> rmse(x - y)
[1] 2.727636