Multiple Regression, Clearly Explained!!!

[@statquest]

Support StatQuest by buying my book The StatQuest Illustrated Guide to Machine Learning or a Study Guide or Merch!!! statquest.org/statquest-store/

[@jolenegan]

1 million subscribers and John still takes the time to respond to every comment on his videos. Thank you for your help with understanding my assignments!

[@statquest]

Thank you!

[@JinSakai-h9v]

Universities should just play your course videos in classes instead of making failed attempts in teaching

[@statquest]

:)

[@peace822]

True😂😂😂😂

[@Pramodi1123]

@@JinSakai-h9v exactly lol

[@meri_mory]

I'm doing a class presentation on this in graduate school so thank you so much for the help! 😭

[@statquest]

Good luck!!

[@Motivations731]

love the learning method😊😊

[@statquest]

Thanks! 😃

[@Mostusernamesaretakn]

We need a StatQuest of the FWL theorem! 🙏🏻

[@statquest]

Noted

[@Mostusernamesaretakn]

@@statquest thank you Josh, it’s related to multiple regression and the partialling out of regression coefficients - I’ve just never found a good intuitive explanation in video or text.

[@gametech7046]

Please make a statquest on multivariate linear regression

[@statquest]

Umm... this video is about multivariate linear regression. So is there something else you were getting at?

[@gametech7046]

@@statquest Sorry I thought multiple and multivariate are different.

[@Hunter-uz9jw]

lol

[@utku_bambu]

thank you for this also, BAM!

[@statquest]

Double bam! :)

[@eveanjun8730]

真的太感谢了🥲❤️❤️❤️❤️讲的好好

[@statquest]

BAM! :)

[@tipzdaturtle]

This is gold

[@statquest]

Thanks!

[@nagaadiga6188]

That intro😂😂

[@statquest]

:)

[@gilr.7]

Thanks a lot for the video!

[@statquest]

Bam1 :)

[@bkairosh8164]

If we used R squared and p-values to see if the tail length is relevant, what the reason for calculating the F value. What purpose does the F value play here?

[@statquest]

The F-value is how we calculate the p-value.

[@3dward838]

If p_multiple > 3, do we generate several F values comparing to simple regressions for each other variable? eg. multiple regression for body length vs. mouse weight, tail length, wheel time would we have to compare: - simple model of body length vs mouse weight and multiple model of body length vs mouse weight, tail length, wheel time - simple model of body length vs tail length and multiple model of body length vs mouse weight, tail length, wheel time - simple model of body length vs wheel time and multiple model of body length vs mouse weight, tail length, wheel time thanks for another excellent video

[@statquest]

Yes. You don't have to do every permutation, but it is pretty common to calculate p-values for "all vs all-minus-one" for each variable in the model.

[@klaudiakornalska1819]

@@statquest is this describing a process of stepwise regression? Where we test the variables to see, if they have an impact and in the end keep those that did have an impact i.e., significant pa value?

[@statquest]

@@klaudiakornalska1819 I think that stepwise regression is related, but has a more formal approach to testing each variable.

[@yeahwhatever7655]

The intro song!😂

[@statquest]

:)

[@SanjaySharma-ys4wr]

Well explained

[@statquest]

Thank you!

[@parniaahmadi5494]

Hi Josh, Do you have any video about multicollinearity and Variance inflation factor?

[@statquest]

Not yet.

[@TheBlueFluidBreathe]

Bro you should have a video only about F-statistic on regressions, you did good explaining but a bit out of the topic.. thank you very much anyway

[@statquest]

I'm not sure I understand your comment. I have a videos on just the F-statistic and regression: kzbin.info/aero/PLblh5JKOoLUIzaEkCLIUxQFjPIlapw8nU

[@nilanjana1588]

please make a video on linear regression assumption.

[@statquest]

I'll keep that in mind.

[@hy7027]

is there any video for the degree of freedom please? i searched a lot of it but still don’t quite understand why each parameter in regression will “steal” a degree

[@statquest]

Unfortunately I haven't made that video yet.

[@thedownwarddoug6642]

Who else is singing “stat quest”

[@statquest]

I hope a lot of people! :)

[@juliasokolovska9016]

Me. Already in my sleep hahah

[@thedownwarddoug6642]

@@juliasokolovska9016 this may be the least rockstar area of KZbin.

[@juliasokolovska9016]

@@thedownwarddoug6642 looks like that

[@abidhussainwani3028]

What is a dummy variable, sir?

[@statquest]

Dummy variables are what we use to replace discrete values in design matrices. For details on design matrices, see: kzbin.info/www/bejne/hHeYkJWqhMZ2n8k and kzbin.info/www/bejne/eaKveKmtnpJohsU

[@daytonight6848]

Josh if i was ____________________________________________REDACTED_______________________________________ thanks josh.

[@praneethduggirala5160]

what would be the n value there in multivariate - 3?

[@statquest]

It's the number of data points.

[@meghthakkar]

BAM !! ez pz lemon squecy

[@statquest]

:)

[@smssyedmohammedshah]

I felt like I was watching cartoon network lite, not fun when your learning

[@statquest]

Noted