 # t statistic?

Hey all, Question about 3.11.c (Formulate a test of the hypothesis that the population correlation coefficient equals zero and determine whether the hypothesis is rejected at a given level of significance). The t statistic calculation is given as: r * sqrt(n-2) / sqrt(1 - r^2) [r = correlation] Where does this formula come from? How does it relate to the standard t test form of (x*-x)/s ? Thanks, Ari

not important to know where formula came from, esp. now. know the formula, know what it tests, and how to use it… for what purpose. after the exam, open up an econometrics text book and learn the details.

I think the difference between the two formulas is that one is to test the significance of the correlation coefficient ® of two variables. Whereas (x*-x)/s is used to test whether the regression coefficient of a predicted value is significant when compared to the true slope (X).

Thanks Lonely. Re: cpk, I find it much easier to remember and use things when I understand them.