Data with outliers

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

AfonsoRod
Posts: 43
Joined: Wed May 31, 2017 7:37 am

Data with outliers

Postby AfonsoRod » Wed May 15, 2019 9:44 am

Dear all

I have a daily dataset from 2009 to 2018, and all my variables have outliers. I want to use the "Inter Quartile Range" to identify the outliers, and then use the @bounds command to remove them.

My question is: I have to transform my variables into logs, so should I remove the outliers before or after I have transformed my variables into logs?

Thank you for the help

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data with outliers

Postby startz » Wed May 15, 2019 9:51 am

Log is a monotonic transformation. Assuming that your values are all positive, it should make no difference whether you change the sample before or after taking logs.

AfonsoRod
Posts: 43
Joined: Wed May 31, 2017 7:37 am

Re: Data with outliers

Postby AfonsoRod » Wed May 15, 2019 10:01 am

startz wrote:Log is a monotonic transformation. Assuming that your values are all positive, it should make no difference whether you change the sample before or after taking logs.


startz, i thought that too, but I have removed the outliers using the @bounds command, and when I transformed my variables into logs, the presence of outliers were confirmed again.

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data with outliers

Postby startz » Wed May 15, 2019 11:46 am

Not sure what happened, but if you remove outliers and then recompute interquartile range the range will change.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 25 guests