Question about Quantitative Methods, Hypothesis Testing (How to read and get critical values from F Distribution Table))

Question is in italics
Hi guys I am having trouble understanding how to read the F Distribution table (Registered for August 2022 exam). For example a question in the cirriculum states:

Test whether the variance of returns is different before the regulation change versus after the regulation change, using a 5% level of significance.

Im good with calculating the F test statistic, but the problem i am having is how to find the crititcal values as the solution states the critical values are:
" With 418 − 1 = 417 and 418 − 1 = 417 degrees of freedom, the critical values are 0.82512 and 1.21194. "
Please can somebody explain to me how to find these values from the F Distribution table, do we choose the infinity line for both df1 and df2 because that just gives me 1 critical value, or do we use the 120 line for both. It would be helpful if someone can explain the process for choosing the critical values from the F Distribution table in detailed steps.

Thanks
Regards

You need to use two separate tables (or two instances of F.INV in Excel): one where the probability is 2.5% and one where it’s 97.5%.

1 Like

Right, but what degree of freedom do i choose, the 120 one or infinity, and also the tables listed at the end of quant reading only show the F tables for 0.05, 0.025, 0.01 and 0.005. When you mention 97.5% is there some seperate table that Im missing or are you referring to Chi test Table

The tables in the back of the curriculum aren’t sufficient for this problem; apparently they want you to use Excel.

On the real exam, if you need to determine a critical value for an F-test, they’ll give you an excerpt from an appropriate table. If you need 417 degrees of freedom, the table they give you will have 417 or, at least, values close to 417 (e.g., 410 and 420, or 400 and 425).

Also note that CFA Institute has adopted the convention that they’ll put the larger variance in the numerator, so you’ll never get a test statistic less than +1.0, and you need only the larger critical value (here, 1.21194, which they clearly got from Excel).

2 Likes

Thanks a lot that made sense

My pleasure.

1 Like