One-Sided Alternatives(cont) y1=B+Bx;+….+Bxi+t Ho:Bi=0 H1: >0 Fail to reject reject Economics 20- Prof anderson
Economics 20 - Prof. Anderson 11 yi = b0 + b1 xi1 + … + bk xik + ui H0 : bj = 0 H1 : bj > 0 0 c a (1 − a) One-Sided Alternatives (cont) Fail to reject reject
One-sided vs Two-sided o Because the t distribution is symmetric testing H: B0 is straightforward. The critical value is just the negative of before o We can reject the null if the t statistic <-c and if the t statistic than -c then we fail to reject the null For a two-sided test. we set the critical value based on a/2 and reject H: B:*0 if the absolute value of the t statistic > c Economics 20- Prof anderson
Economics 20 - Prof. Anderson 12 One-sided vs Two-sided Because the t distribution is symmetric, testing H1 : bj < 0 is straightforward. The critical value is just the negative of before We can reject the null if the t statistic < –c, and if the t statistic > than –c then we fail to reject the null For a two-sided test, we set the critical value based on a/2 and reject H1 : bj 0 if the absolute value of the t statistic > c