Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: Re: FW: ipolate and epolate - positive and integerDV


From   Robert A Yaffee <bob.yaffee@nyu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Re: FW: ipolate and epolate - positive and integerDV
Date   Tue, 05 Sep 2006 08:20:37 -0400

Samira,
     It might be possible to assess excess relative risk
rather than relative risk.  You would add a one to the
relative risk before taking the natural log.  If you
have to negative integers, this might solve your problem.
If you have negative integers less than -1, the problem
would remain unless you rescaled your measure.
     - Bob Yaffee


Robert A. Yaffee, Ph.D.
Research Professor
Shirley M. Ehrenkranz
School of Social Work
New York University

home address:
Apt 19-W
2100 Linwood Ave.
Fort Lee, NJ
07024-3171
Phone: 201-242-3824
Fax: 201-242-3825
yaffee@nyu.edu

----- Original Message -----
From: Nick Cox <n.j.cox@durham.ac.uk>
Date: Monday, September 4, 2006 1:34 pm
Subject: st: RE: Re: FW: ipolate and epolate - positive and integer DV

> I don't know what you are doing downstream 
> of this, but some commands will accept non-integers
> even when the variables "should be" integers. 
> 
> Nick 
> n.j.cox@durham.ac.uk 
> 
> Samira Reis
> 
> > Thank you for your suggestion Nick!
> > 
> > I had done the "logging, interpolating on the
> > logarithmic scale and then antilogging" before
> > following what you had suggested to another person in
> > the statalist some time ago. However, I thought it
> > couldn't work for me because after doing that, values
> > were not integer anymore, and my dependent variable is
> > a count.
> > 
> > But now you said also that decimals values may be not
> > such a problem, that is, "you would be less worried
> > about integer values yourself" then I will probably
> > round the interpolated values to integer.
> > 
> > Thank you very much!
> > 
> > Samira.
> > 
> > 
> > Samira Reis had difficulty sending this to the list.
> > 
> > My own suggestion -- so long as all values are
> > positive --
> > is that Samira should consider logging, interpolating
> > on the logarithmic scale and then antilogging.
> > 
> > Or take square roots -- that doesn't ensure positive
> > values either, but there are several grounds for
> > regarding square roots as "first aid" (Tukey) for
> > count.
> > 
> > I would be less worried about integer values myself.
> > 
> > Nick
> > n.j.cox@durham.ac.uk
> > 
> > Samira Reis [mailto:samira_dias@yahoo.com.br]
> > 
> > Hi Stata users,
> > 
> > I have a data set from 1980 to 2003 with some missing
> > values for both
> > dependent and independent variables. I tried using
> > ipolate command in
> > stata to get values and I used:
> > 
> > by reg: ipolate ent year, gen (ient) epolate.
> > 
> > The problem was that I obtained quite a few negative
> > values and my
> > variables can not be negative and my dependent
> > variable must be integer
> > as well. Could anybody help me to find out a
> > correct way to deal with that, please?
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index