 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: AW: AW: table no obs and missings
Martin, thanks for this quick clarification.
r(F) >= .25 is/was more a test-case. I realise now that there is no good indicator with -oneway- to judge automatically if it is worthwhile looking at a certain table in more detail. So .oneway x y, tab will be our friend ;-)
Marc
-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Martin Weiss
Gesendet: Freitag, 12. Februar 2010 09:53
An: [email protected]
Betreff: st: AW: table no obs and missings
<>
The missing option pertains to the statistics within the -table- being missing, not the values of the row variable:
*************
webuse apple, clear
replace treatment=.z in 1/5
oneway weight treatment
table treatment if r(F)>= .25, c(m we sd we med we)
*************
Do you really want to condition on your F-statistic being more than 0.25?
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Kaulisch, Marc
Gesendet: Freitag, 12. Februar 2010 09:46
An: [email protected]
Betreff: st: table no obs and missings
Dear Statalisters,
Using Stata 11 MP 64bit (Win Server 2008R2) I am confused about some functionality of -table-.
I use the following code:
. oneway `var' fachuhdfak, means scheffe
. table fachuhdfak if r(F)>= .25, c(m `var' sd `var' med `var')
. ttest `var', by(muhd)
. table muhd if r(p) <= .05, c(m `var' sd `var' med `var')
The first -table- command line reports also missing values of variable fachuhdfak - although missing as an option is not specified. Do I misinterpret the missing option?
Table looks like:
X1
X2
X3
.n
.z
The second -table- command line produces the error no observatios (r(2000)) as r(p) is higher than .05. I wonder why -tab2- is ignoring that there are no observations but -table- produces an error?
Any ideas/hints/clarifications?
Marc
iFQ
Institut für Forschungsinformation und Qualitätssicherung Godesberger Allee 90
53175 Bonn, Germany
Tel.: *49 - 228 - 9 72 73 - 25
Fax: *49 - 228 - 9 72 73 - 49
E-mail: [email protected]
<blocked::mailto:[email protected]>
www.forschungsinfo.de
*
* 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/