# st: missing p-value from t-test

 From Ngoc Anh Vo Thi To Statalist Subject st: missing p-value from t-test Date Fri, 23 May 2008 02:16:48 -0700 (PDT)

```Dear  stata-users,

My appologies for a beginner’s question.  I’m estimating the following model using OLS:
lnROA=alpha + beta(1)*lnW(1) + beta(2)*lnW(2) + beta(3)*lnW(3) + gamma*D + epsilon.

Then I calculate an E statistic that equals the sum of beta(1), beta(2), and beta(3). I want to test whether E=0, using a t-test:
ttest  E==0

One-sample t test
------------------------------------------------------------------------------
Variable |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
E_CZE_B |      38   -.0550183           0           0   -.0550183   -.0550183
------------------------------------------------------------------------------
mean = mean(E_CZE_B)                             t =        .
Ho: mean = 0                                 degrees of freedom =       37

Ha: mean < 0                 Ha: mean != 0                 Ha: mean > 0
Pr(T < t) =      .         Pr(|T| > |t|) =      .       Pr(T > t) =      .

I would like to know why all the p-values are missing here? Is it because E is a fixed value for all observations? If this is the case, how can I test this hypothesis, i.e. E=0?

Any help would be very much appreciated.
Anh

*
*   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/
```