The F Test, Comparing Two Variances, Problem 1

[@tumishoramodisa9602]

This guy explains a 2 hour lecture in only 8 minutes!🔥

[@dmcguckian]

Thanks 😊

[@SeVeN-os3jl]

Because tomorrow is the exam, and we have no other option left other than to mug that up .. JK .... He explained greatly

[@AdamJung-m9n]

Thank you so much! This was explained so well and it helped me figure out how to compare two sets of data for my graduate thesis.

[@arshk3322]

Thank you so much. I couldn’t find practice problems on F test anywhere else on KZbin. And you explained it so well.

[@dmcguckian]

You're very welcome!

[@jaweriawahid6902]

Finally a helpful vedio. Thank you Sir 💝

[@dmcguckian]

You’re welcome!

[@nevilleatakere2203]

You are the best!

[@dmcguckian]

Thanks 😊

[@ghalibhassan7047]

Thanks a lot sir. It helped a great deal.

[@zyasmine5649]

Thank you so much

[@dmcguckian]

You’re welcome

[@fabiocaldeira5934]

Do i always put the larger variance on top, if so how do i know which of the two populations will be the first degree of freedom. I use the the excel function to calculate these values

[@alicezhou6252]

Hi, just wanna ask why it's a two sided test but we only calculate one critical value? it's that (2.462) the p-value?

[@dmcguckian]

Hi Alice, we purposely organized our test stat and the claim to ensure that the largest sample variance would belong to the numerator of the test stat. This ensures that the test stat will be greater than one, so we only need to consider whether it is significantly large. This removes the need for the left critical value. The reason we do this is that it is a little difficult to get the lower critical value using our table (it’s not too difficult but we can avoid the need if we just make our sample with the largest sigma our first group).

[@JitendraSingh-vo2xs]

excellent explaination. Thanks

[@aya20008]

What is the non-parametric equivalent to f test( violation of normality distribution and outliers), given that I have two groups? Thanks in advance.

[@bashatujuba5490]

thank you sir it helped a great deal

[@ianrice9083]

Super helpful video but why do you draw your 5's low to high? I have a friend that does low to high for everything but I only noticed you do it for 5's. Just wondering

[@dmcguckian]

I actually didn’t even realize I wrote them in a different way until I used a math app for mental exercises. The app wouldn’t take my answers whenever I had to write a five on the screen with my finger, so I tried writing an s whenever the answers had a 5. The app took the s as a 5, but not my fives no matter how carefully I wrote them. That is how I realized that I was writing them in the opposite order of most people. Believe it or not, none of my students in classrooms that I taught in ever mentioned it. Also, none of my elementary school teachers did either. I can say the normal way of writing them is better. It is much easier to write a proper five doing it the way everyone else does it.

[@ianrice9083]

@@dmcguckian That's a super cool story. Thanks! Keep up the good work!

[@zizojaezekeom3565]

Excuse me, Why didn't you multiply 13.12^2 with 16 and divide it by 15, same thing for the other distribution?

[@acromegaman79]

Where does minus one come from when determining degrees of freedom?

[@dmcguckian]

The best answer is that it is derived from the proof relating to the t-distribution; however, if you want an intuitive way to think about it: for a given sample mean, you could randomly choose at most n-1 values and still have the n values produce the given mean. For example, if four values add to 400, the mean for that sample is 100 (100=400/4). You could randomly replace 3 of the four values with 0 and still end up with an average of 100 by making the last value 400, but you can’t randomly replace all 4 values and guarantee the same average. The last value needs to be selected to ensure the average remains 100, so we are free to choose any value for at most n-1 of the n values, but at least 1 value cannot be randomly generated if we want to still have the same mean.

[@szymonrudnicki3045]

Shouldnt we also find a left critical value on the graph?

[@dmcguckian]

The f-table is a pain to use for the left critical value, so that is the reason we put the larger sample variance on top of the fraction. This ensures that we don't need to look at the left-sided critical value. If you want to find the left-sided critical value, you can visit my website and open the pdf notes for section 9.11. On the last page of those notes, you'll find a description of how to find the left critical value.

[@korirerick5066]

Make a video on when the sample size is large n>40 please

[@LwAcademia]

Thank you so much🔥

[@jacklim258]

How do we know if it is a two-tailed test?

[@dmcguckian]

The problem should say to test a claim that the variances are equal or it should say to test the claim that they are not equal. In either of those scenarios, the test is two-tailed.

[@dinesshr.t1312]

Thanks for your video sir I had a clarification.

[@dmcguckian]

You’re welcome

[@poisoned_durian8]

thank youuuu!!

[@amaliarios78]

Very nice video

[@dmcguckian]

Thank you

[@hafsarizwanahmed5944]

Thanks man 👍

[@dmcguckian]

You’re welcome

[@relaxingmind34]

Sir you devided alfa by 2

[@hafizaayesha6744]

Can we take the sequre of variance?

[@dmcguckian]

No, if you have the variances, you can just divide them to get your test stat.

[@RacineDavid-op6vr]

Why are you dividing the significance level by 2??

[@semaatalan]

I thing it is not correct. If two tailed ve should dividing the significance level by 2

[@UmarKhan-yg9wp]

why is it a two-tailed test ?

[@llaaaza1911]

Because of the equality. All the inequalities are one tail because they are either greater than or more than. The equality is two tailed because it has a boundary on the left (less than) and on the right (more than)

[@siddhantjha7320]

For this question he has used one tailed test right ??​@@llaaaza1911

[@robscovell5951]

I used that calculator at school 40 years ago

[@nafiziqbal814]

Thanks

[@fahadimran6961]

thankyou sir

[@dmcguckian]

You’re welcome

[@magdalenemark2757]

Thanks sir

[@dmcguckian]

You're welcome

[@pratikrai7263]

Is it same if the question given is can we conclude there is difference in the variances

[@dmcguckian]

It sounds like they are asking you to test a claim that two variances are unequal. That should be a two-tailed F test.

[@Mr.BossBaby]

This is a right side tho not equal

[@nurlatifahmohdnor8939]

no-ta = note ka-ki = foot 20 Murray, C. A. (1984)

[@nurlatifahmohdnor8939]

Me-ra' slang. = Im-ra'ah Meran = Am-ran or maybe Im-ran Our relatives in Lu-rah Bi-lut. My mother's cousin. Her grandmother's daughter's son. She married 2 times in her life. Both men are from the same village. K. L.

[@korirerick5066]

I understood this like never

[@mugithmugeez6780]

the voice is very low

[@dmcguckian]

It seems fine on my end. Sorry it’s not loud enough for you. Maybe try headphones.

[@mugithmugeez6780]

@@dmcguckian thanks

[@alvesphiri3938]

T-test ??????

[@JhonnyStone100]

🐐

[@dmcguckian]

:)

[@tadiwanashekerste]

❤❤❤

[@gulhayoraxmonova3460]

thank you very much

[@dmcguckian]

You’re welcome