Statalist The Stata Listserver


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

st: RE: RE: RE: Re: RE: RE: Continous to categorical variable


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: RE: Re: RE: RE: Continous to categorical variable
Date   Wed, 21 Feb 2007 17:55:39 -0000

Jason's idea is easier than he implies. You 
can go 

egen std_mpg = std(mpg)

and then use any rounding you like, e.g. 

gen class_mpg = round(std_mpg)
gen class_mpg = floor(std_mpg)
gen class_mpg = ceil(std_mpg) 

Recoding to integers 1 up seems superfluous 
as the integer codes have a meaning already. 

In the case of income, working with log(income)
or quantiles would seem more likely to be 
helpful. 

Nick 
n.j.cox@durham.ac.uk 

Jason Yackee
 
> I would use standard deviations to create the categorical 
> variable.  Say you wanted six categories -- "1" is greater 
> than negative 2 stdev from the the mean, "2" is any 
> observation between negative 1 and negative 2 stdev from the 
> mean, 3 is between 0 and negative 1 stdevs, 4 is between 0 
> and positive 1 stdevs, and so on.
> 
> The only thing "by hand" is calculating the descriptive 
> statistics, and recoding the variable six times.
 
Mentzakis, Emmanouil

> > I do not want the variable to "become more like a normal
> distribution".
> > What I would like is the categories created to be such that 
> the tails
> > contain less individuals, with an increase in the numbers as we get
> > closer to the middle category.

Mentzakis, Emmanouil 

> >> I have a continous variable (i.e. income) and I would like to
> >> transform it into a categorical one (e.g 5 categories/levels or
> more).
> >>
> >> I would like to ask if there is any way that I can ask stata to
> create
> >
> >> this variable deciding the appropriate cut-off points automatically
> so
> >
> >> that the categories follow aproximately a normal 
> distrubution or they
> >> are of equal size.
> >
> > For the latter have a look at -help egen- and look at the cut
> function.
> > For the former: how would you expect a variable to become 
> more like a
> > normal distribution by making it coarser?

*
*   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