I have this problem on Schweser.
The critical F-value for testing H0=b1=b2=b3=b4=0 vs. Ha: at least one bj is not equal to 0 at the significance level 5%. The # of observations are 65.
So I go back to F-Table to find the critical value. The # of independent variables which is 4 we can find from the top row of F-Table, and the # of observations we can find from the side row of F-Table, so the value is 2.53. That’s the answer.
My question is that what is the role of 5% significance level? It’s got nothing to do to find the answer. Right? We don’t really need the Significance level when we find F critical value or do the F-Test. Right?
Every F table is based on a significance level, represented by the area to the right of the curve.
To find the critical F value, you need the significance level (e.g 5%) and the degree of freedom of the regression (numerator) and residual (denominator).
Note that in the back of the Level II Quant book there are several F-tables. If you look in the header for each table, it will tell you the area in the right tail; that’s the significance level. Each table has one, and they’re all different from each other.
When finding the critical value for the F-test, I recall from L1 testing variances that you divide the significance level in half when looking at the tables when testing if a value is equal to 0 then use the appropriate df for both numerator and denominator (α/2). I’m noticing the L2 book for the 2022 curriculum doesn’t do that when you’re doing a two tailed test when testing slope coefficients. You use just α. Can someone explain why we’re doing the F-test differently here?