There is more than one way to do it.
Here is an alternative.
gen Variable = ""
gen dfuller = ""
local i = 0
foreach v in m1 m2 consump {
local ++i
replace Variable = "`v'" in `i'
dfuller `v'
replace dfuller = string(r(Zt), "%6.4g") in `i'
}
tabdisp Variable, cell(dfuller) in 1/`i'
Disadvantages of this:
1. The results are not aligned with your existing data. Could
lead to misunderstanding.
2. Will be problematic if #results > _N.
3. -tabdisp- is limited to 5 cell variables.
Advantages of this:
1. Keeps results together with existing data.
2. Further analysis, graphing, etc., often easy.
3. Can control formats desired at source.
4. Can elaborate, e.g. by using variable labels.
Nick
[email protected]
Austin Nichols
> Partha Deb <[email protected]>:
> I think you can trick -estout- into doing that, by generating
> variables holding the values of your test statistics, and using its
> summary stats capability. But that would be horribly inefficient
> compared to just writing a tab-delimited text file using the -file-
> command:
>
> webuse friedman2, clear
> file close _all
> file open t using /test.txt, write replace
> file write t "Variable" _tab "dfuller"
> qui dfuller m1
> file write t _n "m1" _tab "`: di %6.4g r(Zt)'"
> qui dfuller m2
> file write t _n "m2" _tab "`: di %6.4g r(Zt)'"
> qui dfuller consump
> file write t _n "consump" _tab "`: di %6.4g r(Zt)'"
> file close _all
> type /test.txt
>
> On 9/4/07, Partha Deb <[email protected]> wrote:
> > Hi,
> >
> > Is there a package akin to -estout- or -outreg- to make
> tables from test
> > statistics? I'm conducting a series of tests on a number
> of variables and
> > would like to tabulate the statistics. E.g., from
*
* 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/
