Question

Question: How would you scan for outliers in your dataset? What would you do with data points that are considered outliers?

Answer 1

To identify outliers there are many methods--
a) Suppose you have single data set, then draw box plot and use    criteria to detect outliers
b) We have data on two variable x and y, then simple draw scatter plot.The value which appears most extreme is considered as outliers.

c) We can also use Z -score to detect outlier .

What to do with outliers:

a) Drop the outliers from the data set
b) Cap your outlier data
c) We can assign a new value to outlier within the data set.
d) Try some transformation.

( Please give thumps up, if you like my answer )