Outlier_problems


#1

What is the solution when we have more than 50% outliers in the data?
How we analyze that data without removing outliers(information)?
Is there any way to solve the problem of outliers?
What can we do on this data?


#2

Hi @sknadar1

50% are not outliers, 1% yes !!! I do not know your problem but I guess it is to do with states of the “system” you want to model. Check about hierarchical models or mixed models. If you give more details about the type of problem you want to solve I am sure many people could help you.

Best regards

Alain


#3

Try to apply log or scale the data. In that way you can analyze the data without removing any of them.


#4

How did you find 50% of your data are outliers? can you share the box plot