Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: RE: Summary statistics table problem


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: Summary statistics table problem
Date   Wed, 29 Oct 2008 13:15:57 -0000

Changing the notation, George seems to be asking this: 

1. He has variables x1, x2, up that he wants on the _rows_ of a table. 

2. He has a variable y whose categories he wants as the _columns_ of a
table. 

3. He wants the means of each x given y inside the table. 

If so, 

I suspect this would be a lot easier if the x's were -reshape long-ed
into one. 

I wouldn't try to program it as a table problem. It's more a matter of
data management upstream of the tabulation. 

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

George Hoffman

In Martin's solution, the x is just a dummy.
The rows are the contents - means of the var1 .. Varn - for each value
of x,
which in this case is just a single dummy entry.
The columns are the values of rep78 - which are labels by their values
Thanks for your thoughts - I'm thinking there is already a way to do
this
that I'm somehow overlooking, because it seems like a common function.

Matt Spittal
 
> Adding labels to the table is no problem.  You just need to define
what the
> labels are first with the -label define- and -label values- commands.
Here is
> an illustration:
> 
> // Martin's code
> sysuse auto, clear
> generate x = 1
> 
> // add labels
> label define rep78 1 "Very good" 2 "Good" 3 "Average" 4 "Poor" 5 "Very
poor"
> label values rep78 rep78
> table x rep78, contents(mean price mean mpg)

George Hoffman

> Yes, this works, except for no row labels, but I can live with this.
> I might be inspired to write something to 'fix' this problem.

Martin Weiss
 
>> Even dirtier:
>> 
>> *********
>> sysuse auto, clear
>> g x=1
>> table x rep78, contents(mean price mean mpg )
>> *********

George Hoffman

>> thanks, but not quite.
>> Tabstat var1 var2 ... Varn, stat(mean) by(byvar) either:
>> col(var): yields columns of the var1, var2, with the by(byvar) as
rows
>> Col(stat): yields sequential single columns of output for each
var1..varn
>> 
>> I'm looking for the by(byvar) to appear in columns side by side, with
rows
>> for var1...varn

Martin Weiss
 
>>> -tabstat var1 var2, statistics( mean ) by(Hour6) nototal
>> columns(variables)-
>>> ?

George Hoffman
 
>>> This should be simple, but I can't find a straightforward way to do
this.
>>> Xt  dataset, with hours 1-48 for time, also summarized by hour6
(1-6,
>>> 7-12,...)
>>> Many row variables.
>>> 
>>> I want a table of means by hour6:
>>> 
>>> Var         Hour6
>>>             1       2       3       4       5       6       7
8
>>> var1        50      55      55      57      60      62      61
57
>>> var2        47      34      23      12      6       5       7
6
>>> Etc

*
*   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/



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