Statisticians, number magicians,

I was wondering what the best way is to check for the equality of two parameters including possibly a confidence interval and p-value.

$$H_0:\beta_1=\beta_2\vert\ \tilde{y}_1=\hat{\beta}_1\tilde{x}_1;\tilde{y}_2=\hat{\beta}_2\tilde{x}_2;n_1\not=n_2$$

Thanks, Rik

  • $\begingroup$ There's not enough information here. What is known about the $x$ and $y$ variables? Is there really no error here (so that a single observed $x_1$ and $y_1$ will tell us $\beta_1$ exactly)? $\endgroup$ – Glen_b -Reinstate Monica Apr 14 '15 at 1:45
  • $\begingroup$ Sorry they are not single observations, they are two different populations; so it is two regressions with no constant. I'm not sure what the right notation is. $x_1$ and $y_1$ are vectors... $\endgroup$ – Rik Apr 14 '15 at 1:51
  • $\begingroup$ Are you testing equality of regression coefficients? $\endgroup$ – SmallChess Apr 14 '15 at 1:53
  • $\begingroup$ Yes, but without a constant, only a slope... $\endgroup$ – Rik Apr 14 '15 at 1:54
  • $\begingroup$ Isn't this just like the text-book method but with the constant be zero? $\endgroup$ – SmallChess Apr 14 '15 at 1:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.