sounds like a problem with you're getting the degrees of
freedom correct -- what is the N for the regression and
how many parameters are in the regression; when you used
a table, how many df did you use in the lookup?
Rich
sara borelli wrote:
Dear Statalist members,
I have a problem in using the cluster option. After I
use cluster, I get different interpretations of the
significance of the results using the t-statistic or
the p-value. If I omit cluster, then the t-statistics
and p-value give the same answer, I also calculate the
p-value by "hand" with the statistical tables and the
results correspond to the output in stata.
But when I use cluster, the p-values look "wrong",
that is if I calculate the p-value using the
"new-clustered" t-statistics and the statistical
tables I get a result that is different from the stata
output
example:regression NO CLUSTER
coeff std err t p-value
variable x: -.0156935 .0091102 -1.72 0.085
confidence interval
-.0335624 .0021754
p-value and t give the same answer and p-value is
right according to tables
example:same regression with CLUSTER
coeff std err t p-value
variable x: -.0156935 .0073801 -2.13 0.053
confidence interval
-.0316372 .0002501
p-value and t give different answers, according to the
statistical tables the p-value should be 0.033
I would appreciate any help
Sara Borelli
__________________________________________________
Do You Yahoo!?
Poco spazio e tanto spam? Yahoo! Mail ti protegge dallo spam e ti da tanto spazio gratuito per i tuoi file e i messaggi
http://mail.yahoo.it
*
* 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/
*
* 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/