Question

What is the difference between a suspicious data point and an extreme data point?

Answer 1

Suspicious data point: when a data points is observed as abnormal in any kind of aspect , then it is called as Suspicious data point.

Extreme data point: when a data point is observed far away from the tendency of the data it is called as extreme data point

Example: let us consider a data set which is having time taken for door deliveries of an online food delivery person.

let us assume in general in and around 5 deliveries per hour completed by the person.

but in the data if suddenly one observation is recorded as 35 deliveries per hour, then it is called as suspicious data point because 35 deliveries per hour is almost impossible. it may be wrongly entered.

where as if any observation is recorded as 15 deliveries per hour, it is considered as extreme data point because sometimes 15 deliveries per hour is possible. but it is very far away from the general observations which are 5 deliveries per hour.

For the data analysis purpose we can remove a suspicious data point and continue the analysis, because it may not be belongs to the data set.

But we until unless we have a strong evidence, we should not remove the extreme data point from the data set. because even it is very far away from the tendency of the data, still it is the part of the data set.

try to understand this for different situations as your concern.