ESS210B
Prof. Jin-Yi Yu
Analysis of Variance (ANOVA)
qWe then use F-statistics to test the ratio of the variance explained by the regression and the variance not explained by the regression:
q                       F = (b2Sx2/1) / (Se2/(N-2))
qSelect a X% confidence level
qH0: b = 0
q            (i.e., variation in y is not explained by the linear regression but
q            rather by chance or fluctuations)
q     H1: b ¹ 0
qReject the null hypothesis at the a significance level if F>Fa(1, N-2)
regression slope in population