Usually I am determining if a regressand is statistically different than 0. How do I determine if it is statistically different than 1?

Thanks.

Printable View

- Dec 10th 2010, 08:25 PMdwsmithStatistically different than 1
Usually I am determining if a regressand is statistically different than 0. How do I determine if it is statistically different than 1?

Thanks. - Dec 13th 2010, 05:09 PMdwsmith
Testing statistically different than uses the same t-critical values as two tailed test but the t-value is found by:

$\displaystyle \displaystyle t=\frac{\hat{\beta}-1}{se(\hat{\beta})}\Rightarrow \begin{cases}

|t|>t-crit, & \mbox{reject the null that}\ \hat{\beta}=1 \\

|t|\leq t-crit , & \mbox{fail to reject the null}

\end{cases}$

If you want to find out if it is statistically different than -1, you add 1 to beta hat.