Le 04/11/2004 14:54, Herve STOLOWY a �crit :
Dear Richard and all of you who replied:
I now better understand my error of interpretation (more related to stats
than to Stata !!!). The p-value of the coefficient 2.02 is 0.056 (i.e.,
not significant at the 0.05 level).
The number of df is: 3 (model) + 22 (residual) = 25 (total).
Do you know where I could find the critical value corresponding to each
df, and particularly this one?
Best regards
Herv�
di invttail(df,0.5*alpha)
will give you the (absolute value of the) critical value for a test of
Beta= 0 at the alpha % level for the specified df.
hence,
di invttail(25,0.025)
gives the critical value at the 5% level for 25 df (which happens to be
2.0595386)
Except the d.f. in this case = 22, so the critical value is 2.0738731.
