Multiple Regression, Clearly Explained!!!

This StatQuest shows how the exact same principles from "simple" linear regression also apply multiple regression. At the end, I show how to test if a multiple regression is better than a simple regression.
If you'd like to support StatQuest, please consider...
Patreon: / statquest
...or...
UA-cam Membership: / @statquest
...buy my book, a study guide, a t-shirt or hoodie, or a song from the StatQuest store...
statquest.org/statquest-store/
...or just donating to StatQuest!
www.paypal.me/statquest
Lastly, if you want to keep up with me as I research and create new StatQuests, follow me on twitter:
/ joshuastarmer
Correction:
2:49 I left off some of the parentheses for the equation for F. The numerator should be: (SS(mean) - SS(fit))/(pfit - pmean)
#statquest #regression

[@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!

[@user-lc9ho9xi3m]

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

[@statquest]

:)

[@peace822]

True😂😂😂😂

[@meri_mory]

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

[@statquest]

Good luck!!

[@eveanjun8730]

真的太感谢了🥲❤️❤️❤️❤️讲的好好

[@statquest]

BAM! :)

[@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.

[@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.

[@utku_bambu]

thank you for this also, BAM!

[@statquest]

Double bam! :)

[@parniaahmadi5494]

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

[@statquest]

Not yet.

[@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.

[@gilr.7]

Thanks a lot for the video!

[@statquest]

Bam1 :)

[@tipzdaturtle]

This is gold

[@statquest]

Thanks!

[@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

[@praneethduggirala5160]

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

[@statquest]

It's the number of data points.

[@SanjaySharma-ys4wr]

Well explained

[@statquest]

Thank you!

[@nagaadiga6188]

That intro😂😂

[@statquest]

:)

[@nilanjana1588]

please make a video on linear regression assumption.

[@statquest]

I'll keep that in mind.

[@yeahwhatever7655]

The intro song!😂

[@statquest]

:)

[@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: ua-cam.com/video/NF5_btOaCig/v-deo.html and ua-cam.com/video/CqLGvwi-5Pc/v-deo.html

[@meghthakkar]

BAM !! ez pz lemon squecy

[@statquest]

:)

[@daytonight6848]

Josh if i was ____________________________________________REDACTED_______________________________________
thanks josh.

[@smssyedmohammedshah]

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

[@statquest]

Noted

[@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 UA-cam.

[@juliasokolovska9016]

@@thedownwarddoug6642 looks like that