Page 22 - FINAL CFA II SLIDES JUNE 2019 DAY 2

MODULE 7.5: ANOVA TABLES, R , AND SEE
2

• The null and alternative hypotheses are: H : b = 0 versus H : b ≠ 0.
a
0
1
1
• The critical F-value for 1 and 34 df at a 5% significance level is approximately 4.1.
• Remember, it’s a one-tail test, so we use the 5% F-table!

• Reject the Ho and conclude that the slope coefficient is significantly different than zero.

• Recall from the earlier examples we also rejected the Ho using the t-statistic and that the 95% confidence interval did not
include 0.

• Note that here, t = 2.46 ≈ F = 6.33. The difference is due to rounding of the t-statistic to two decimal places.
2
2
Note: The F-test is not as useful when we only have one independent variable because it tells us the same thing as the t-test of
the slope coefficient. Make sure you know that fact for the exam, and then concentrate on the application of the F-test
in multiple regression.