Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
<<winmail.dat>>
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: How to avoid negative fitted values
From
"Craig, Benjamin M." <[email protected]>
To
<[email protected]>
Subject
st: RE: How to avoid negative fitted values
Date
Fri, 20 Aug 2010 23:18:09 -0400
There are two ways that make sense to me, but their use depends on your theory.
1. glm with a log link will produce positive estimates (see work of manning and mullahy)
2. Inequality constraints on the parameter space (much more difficult)
It sound like you are using a linear model on skewed data, and you are overshooting the skew by assuming a symmetric distribution. Maybe if you assume a gamma distribution (see Basu) your predictions will get better. Again this is all dependent on your theory about the origin of the error. I am not in real estate, sorry.
Ben
Benjamin M. Craig, Ph.D. Assistant Member, Health Outcomes & Behavior Moffitt Cancer Center Associate Professor, Department of Economics University of South Florida Contact Information 12902 Magnolia Drive, MRC-CANCONT Tampa, FL 33612-9416 Phone: (813) 745-6710 Fax: (813) 745-6525 [email protected]
________________________________
From: [email protected] on behalf of Laurie Molina
Sent: Fri 8/20/2010 10:26 PM
To: [email protected]
Subject: st: How to avoid negative fitted values
Dear statalist,
I am triying to imput rental prices for houses to a sample using some
characteristics of the construction which are shared between my data
and the data in another database in which rental prices are declared.
When i use OLS i get negative fitted values both in the original
database as in my database.
How can ensure that the predicted values will be non negative?
Thank you very much in advance.
Laurie
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
This transmission may be confidential or protected from disclosure and is only for review and use by the intended recipient. Access by anyone else is unauthorized. Any unauthorized reader is hereby notified that any review, use, dissemination, disclosure or copying of this information, or any act or omission taken in reliance on it, is prohibited and may be unlawful. If you received this transmission in error, please notify the sender immediately. Thank you.