How to Decide Which technique needs to use for Missing Value Treatment in R. For example where to use Mean and Median imputation, where to use Knn Imputation and so on.
Mean is more sensitive to outliers as compared to the median. So, if the distribution of a numerical variable is highly skewed imputing missing data using the mean is probably not a good idea. In such cases imputing using median is preferred.
And KNN imputation is used when it is assumed that a point value can be approximated by the values of its nearby points.