9:51 a.m.
Hi,
I'm looking at panel regression, and one of the assumptions in regression is to remove
outliers. My data has a lot of outliers, and it impacts my residuals which are definitely
not normal. I'm wondering if there is an easy way to remove the outliers?
Thank you
Alison
Alison Loddick
BSc MSc PGCE CStat
Learning Development Tutor (Mathematics and Statistics)
Library and Learning Services
DDI +44 (0)1604 893502
[https://static.hobsons.co.uk/northampton/usermedia/Staff%20News/2017/New%...]
northampton.ac.uk
University of Northampton, Waterside Campus,
University Road, Northampton, NN1 5PH United Kingdom
University of Northampton: Transforming Lives and Inspiring Change www.northampton.ac.uk
This e-mail is private and may be confidential and is for the intended recipient only. If
you are not the intended recipient you are strictly prohibited from using, printing,
copying, distributing or disseminating this e-mail or any information contained in it. We
virus scan all E-mails leaving The University of Northampton but no warranty is given that
this E-mail and any attachments are virus free. You should undertake your own virus
checking. The right to monitor E-mail communications through our networks is reserved by
us.
Disclaimer
The information contained in this communication from the sender is confidential. It is
intended solely for use by the recipient and others authorized to receive it. If you are
not the recipient, you are hereby notified that any disclosure, copying, distribution or
taking action in relation of the contents of this information is strictly prohibited and
may be unlawful.
This email has been scanned for viruses and malware, and may have been automatically
archived by Mimecast, a leader in email security and cyber resilience. Mimecast integrates
email defenses with brand protection, security awareness training, web security,
compliance and other essential capabilities. Mimecast helps protect large and small
organizations from malicious activity, human error and technology failure; and to lead the
movement toward building a more resilient world. To find out more, visit our website.
1:24 p.m.
Am 13.09.2022 um 11:51 schrieb Alison Loddick:
> I’m looking at panel regression, and one of the assumptions in
>
regression is to remove outliers. My data has a lot of outliers, and >
it impacts my residuals which are definitely not normal. I’m > wondering
if there is an easy way to remove the outliers?
Hello, for a standard regression (including the panel case), normality
is not a required assumption. (Unless you want to do "exact" inference
in small samples.) So perhaps you don't need to do anything at all.
Having said that and with respect to your practical question, after
initial estimation you could check for which observations the abs value
of the residuals (which you can save from your initial estimate)
exceeds, say, three times the standard error of the regression (which
you can also save). Then you could either restrict the sample to where
this is not the case, or you add a bunch of corresponding
impulse/observation dummies to your model. Right now I don't have the
time to be very specific about the details, sorry; maybe someone else
can jump in. (The problem with excluding observations in the panel case
is that very quickly the active dataset will lose the official panel
status in gretl because of the gaps. So the dummy solution might be better.)
The "leverage" command would be useful, but unfortunately only works
after the "ols" command, not after "panel". Maybe it could be enabled
at
least for fixed-effects estimation (in a future gretl version).
I'm not aware of any built-in functionality or contributed function
package which automatically constructs outlier-related dummies yet, but
maybe I'm missing something.
cheers
sven
4:24 p.m.
In any firm of regression I hate the idea of removing outliers. They may
contain valuable in. First I would ask if there were errors in Measurement
or coding in the outliers and can the errors be corrected. Are the errors
due to the mistaken use use different units? Are there changes in
technology or other methodologies that might have caused the outliers.
Answering this and similar questions relative to your topic should guide
you on what you should do.
The large number outliers may also indicate that you have a problem with
the model that you are estimating. Have you some omitted variables? Should
you be transforming some of your variables?
Gretl does not provide answers to your questions. Removing outliers is a
last resort.
On Tue 13 Sep 2022, 10:51 Alison Loddick, <Alison.Loddick(a)northampton.ac.uk>
wrote:
Hi,
I’m looking at panel regression, and one of the assumptions in regression
is to remove outliers. My data has a lot of outliers, and it impacts my
residuals which are definitely not normal. I’m wondering if there is an
easy way to remove the outliers?
Thank you
Alison
*Alison Loddick*
*BSc MSc PGCE CStat*
Learning Development Tutor (Mathematics and Statistics)
Library and Learning Services
*DDI *+44 (0)1604 893502
[image:
https://static.hobsons.co.uk/northampton/usermedia/Staff%20News/2017/New%...]
northampton.ac.uk
*University of Northampton, *Waterside Campus,
University Road, Northampton, NN1 5PH United Kingdom
University of Northampton: Transforming Lives and Inspiring Change
www.northampton.ac.uk This e-mail is private and may be confidential and
is for the intended recipient only. If you are not the intended recipient
you are strictly prohibited from using, printing, copying, distributing or
disseminating this e-mail or any information contained in it. We virus scan
all E-mails leaving The University of Northampton but no warranty is given
that this E-mail and any attachments are virus free. You should undertake
your own virus checking. The right to monitor E-mail communications through
our networks is reserved by us.
*Disclaimer*
The information contained in this communication from the sender is
confidential. It is intended solely for use by the recipient and others
authorized to receive it. If you are not the recipient, you are hereby
notified that any disclosure, copying, distribution or taking action in
relation of the contents of this information is strictly prohibited and may
be unlawful.
This email has been scanned for viruses and malware, and may have been
automatically archived by Mimecast, a leader in email security and cyber
resilience. Mimecast integrates email defenses with brand protection,
security awareness training, web security, compliance and other essential
capabilities. Mimecast helps protect large and small organizations from
malicious activity, human error and technology failure; and to lead the
movement toward building a more resilient world. To find out more, visit
our website.
_______________________________________________
Gretl-users mailing list -- gretl-users(a)gretlml.univpm.it
To unsubscribe send an email to gretl-users-leave(a)gretlml.univpm.it
Website:
https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/
590
days inactive
590
days old
2 comments
3 participants
participants (3)
-
Alison Loddick -
John C Frain -
Sven Schreiber