- Using this technique you can replace the extreme values of your data.
- This is just to limit the effect of the outliers on the calculations or the results you get from the data
- The mean value calculated after such replacement of the extreme values is called winsrized mean
- For example, what is 90% winsorization?
- It means the replacement of the top 5% and bottom 5% of the data.
- The top 5% of the data is replaced by the value of the data at the 95th percentile and
- The value of the bottom 5% of the data is replaced by the value of the data at the 5th percentile.
We use the same dataset houseprice.csv
Load the libraries and read data
Descriptive statistics
Separate features and target(price)
Original data means
Draw boxplot using Original Data
Transform the data using Winsorizing Technique
Transformed data
Draw Boxplot using normalized data
Find mean of feature bedrooms
Find mean of Bathrooms
Find mean of sqft_living
Find mean of sqft_lot
Find mean of floors
Find mean of sqft_above
Find mean of sqft_basement
Find mean of sqft_living15
Find mean of yr_built
Normalization avoid the extreme values. So when we compare the means of feature before and after normalization we notice the mean of all features have reduced