wonderfull - but not quite since I want to save the results with
values (or labels) of both variables in the group variable (g).
but when using
svy: mean bpsystol, over(region smsa, nolabel)
mat b=e(b)
mat colnames b= `e(over_labels)'
mat li b
drop _all
svmat b
xpose, clear
gen row = ""
forval o = 1/`c(N)' {
local r : label g `o'
replace row = "`r'" in `o'
}
list
+-----------------+
| v1 row |
|-----------------|
1. | 125.7034 NE 1 |
2. | 127.8939 NE 2 |
3. | 129.4605 NE 4 |
4. | 126.8856 MW 1 |
5. | 126.381 MW 2 |
|-----------------|
6. | 127.6242 MW 4 |
7. | 127.0757 S 1 |
8. | 123.8047 S 2 |
9. | 127.561 S 4 |
10. | 125.6318 W 1 |
|-----------------|
11. | 125.0983 W 2 |
12. | 128.6279 W 4 |
and then extracting the two values gives me the result I wan - a file
with the resulsts of svy: mean .. , over()
| v1 row var1 var2 |
|-------------------------------|
1. | 125.7034 NE 1 NE 1 |
2. | 127.8939 NE 2 NE 2 |
3. | 129.4605 NE 4 NE 4 |
4. | 126.8856 MW 1 MW 1 |
5. | 126.381 MW 2 MW 2 |
+-------------------------------+
I am sure there are more elegant ways to do this, but I cannot get the
more obvious ones to work for reasons I am sure are obvious to adpets.
THanks for all help
Richard
On Thu, Jul 9, 2009 at 4:02 PM, Nick Cox<[email protected]> wrote:
> I stand corrected. See dictionary entries for hubris.
>
> Nick
> [email protected]
>
> Austin Nichols
>
> Richard Palmer-Jones et al.--
>
> Note that -collapse- does not support -svy- for this kind of problem:
> "pweights may not be used with sd, semean,..."
> But the obvious solution is simply to label the columns of a matrix,
> and -egen group, label- before -mean- will give nicer labels:
>
> webuse nhanes2, clear
> egen g=group(region smsa), label
> svy: mean bpsystol, over(g)
> mat b=e(b)
> mat colnames b= `e(over_labels)'
> mat li b
>
>
> On Thu, Jul 9, 2009 at 10:38 AM, Nick Cox<[email protected]> wrote:
>> Just use -collapse- as recommended and this problem, such as it is, becomes irrelevant. As the help tells you, -collapse- supports all kinds of weights. Being clueless is not an excuse for not reading what is provided!
>>
>> Nick
>> [email protected]
>>
>> Richard Palmer-Jones
>>
>> Thanks folks, indieed, but .. it is the values of the _subpops that I
>> want to use - in martyen's case the values of foreign and rep78
>>
>> . mean mpg, over(for rep, nolabel)
>>
>> Mean estimation Number of obs = 69
>>
>> Over: foreign rep78
>> _subpop_1: 0 1
>> _subpop_2: 0 2
>> _subpop_3: 0 3
>> _subpop_4: 0 4
>> _subpop_5: 0 5
>> _subpop_6: 1 3
>> _subpop_7: 1 4
>> _subpop_8: 1 5
>>
>> --------------------------------------------------------------
>> Over | Mean Std. Err. [95% Conf. Interval]
>> -------------+------------------------------------------------
>> mpg |
>> _subpop_1 | 21 3 15.01359 26.98641
>> _subpop_2 | 19.125 1.328768 16.47348 21.77652
>> _subpop_3 | 19 .7862783 17.43101 20.56899
>> _subpop_4 | 18.44444 1.528535 15.3943 21.49459
>> _subpop_5 | 32 2 28.00906 35.99094
>> _subpop_6 | 23.33333 1.452966 20.43398 26.23268
>> _subpop_7 | 24.88889 .9043789 23.08423 26.69355
>> _subpop_8 | 26.33333 3.122499 20.10248 32.56418
>> --------------------------------------------------------------
>>
>>
>> and, I should have said I am using svy: mean for important reasons.
>> collapse does not have a svy equivalent, I think?
>>
>> I cannot see an option on parmest to use the e(over_labels), so, still
>> in the dark I am afraid.
>>
>> Richard
>>
>>
>> On Thu, Jul 9, 2009 at 12:56 PM, Nick Cox<[email protected]> wrote:
>>> Using -collapse- is another (and easier) possibility.
>>>
>>> Nick
>>> [email protected]
>>>
>>> Martin Weiss
>>>
>>> cap ssc inst parmest
>>> sysuse auto, clear
>>> mean mpg, over(for rep, nolabel)
>>> parmest, saving(myfile, replace)
>>> u myfile, clear
>>> drop eq t dof
>>> l
>>>
>>> Richard Palmer-Jones
>>>
>>> I want to compuile the ereturned e(b) and e(v) results of a "mean
>>> var1, over(var2 var3, nolabel)" into a usable form, such as rows in
>>> new variables or a matrix, together with the relevatn values of var2 &
>>> var3
>>>
>>> I think this means making use of the values in e(over_labels), but how
>>> to do so escapes me.
>>
>> *
>> * 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/
>>
>
> *
> * 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/
>
> *
> * 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/
>
*
* 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/
