How to handle outliers

Question

I have data with couple of outliers because of which the predictive model isnt converging, how should I handle such data points.

Answer 1

There are multiple ways to handle outliers and the best technique to do the same isn't defined. You'll have to try different approaches and check which one of them best suits the data. This can be done by fitting a predictive model on the transformed data.

Different approaches can be found on the following link: https://bit.ly/2zYwUMy

Answer 2

You can eiether remove them from your dataset or highlight the others in visualizations to show the outlier characterstics.

There are various ways just like anmol said, you just need to try the best method to make the problem or scenario better to understand.