You can also find this video on his website. Then you can make some questions too so you learn and understand it even better and it provides you videos in the right order! www.khanacademy.org/math/statistics-probability/describing-relationships-quantitative-data/regression-library/v/introduction-to-residuals-and-least-squares-regression
Hi Sal! What program are you using to make those straight line "masks" (where you orient the line "mask" 1st and then draw over it) I'm not seeing this feature in Smoothdraw...I see drawing straight lines but nothing that lets you lay down the line "mask" and then draw so it only fills in inside the "mask" (e.g. 1:46 of this video)
First, THANKS. Second, please resolve 2 issues: 1. Why is the line named Regression Line. From simple English language, I understand the Regression is the opposite of Progression. There should be some thought behind statisticians choosing this word to-draw-a-line-to represent-data. What's that reason? 2. For certain reason, I'm supposed to learn the Partial Least Squares Regression. Importantly, I don't have a background in Statistics. I'm finding it difficult to chart a conceptual-rung-of-ladder to achieve that goal. Can you please guide me to a series of URLs to learn PLS starting from basics through appreciating the concept nicely? This is my need and I'd be grateful if you address it. Thanks and best regards.
why we are squaring the residuals ?? i case of negative values if we upply mod then it will become positive then what the case with the squares of the residuals
I'm not sure but I can see a suspicious moment in this video Your line is approximately y = 5x - 175 if you'll give to ur model the 60 value, it returns 125. Not 150. You've drawn the line that is parallel with axis Y, but it has to be perpendicular for your regression line, doesn't it?
If the aim of the least squares is to elevate the residuals to allow them to have a larger impact in the model, why don't we elevate them by 3, 10, 1000...? That would represent an even higher impact in the model
i still don’t get it :/ you lost me at weight =m(height)+b what does that mean!! please am i like severely mentally i’ll and can’t understand rules or was it not directly explained why we suddenly wrote that like the cause not even why but the why for the why
in less than 8 mins, you taught me what my teacher couldn't in a whole semester. Thank you sooooo much, sensei!
I feel you
Bro same
you spent a whole semester learning this?
@@joeymorse365 haha
Hmmmm. I wonder how much you actually paid attention in class...
Even though these videos are getting fairly old now I still think they are amongst some of the best on the Internet!
Eh, old or not, math is math. It's not changing any time soon. And I agree, these videos are a godsend
What playlist does it belong to?
Ive been struggling with this for my exam tomorrow and this really helped me understand it better- thanks man
did you pass?
@@nicholasneuman1587 a few years late
@@idkagoodname7686 a few months early
As a freshie in HS I’m watching this right before midterms and BOY HAS THIS HELPED ME A BIT
seriously I have been following every single video of yours. such great help. so much easier to understand than the lectures and so much fun!!
iv'e been searching everywhere and watching videos for an unambiguous explanation of OLS. Finally someone explains properly
This helped me understanding linear regression as I am studying machine learning
7:10 The higher the absolute deviation from the line, the heavier the weight, because of squaring
Where is the link to the next video? I suggest you put a link to the next video at the end of the clip.
You can also find this video on his website. Then you can make some questions too so you learn and understand it even better and it provides you videos in the right order!
www.khanacademy.org/math/statistics-probability/describing-relationships-quantitative-data/regression-library/v/introduction-to-residuals-and-least-squares-regression
Always find my answers from here thank you
khanacademy is the G.O.A.T.
you never cease to surprise.
Thankd, it is Simple expression for all levels of students to understand.
You taught me what i needed to know in 6 minutes and my math teacher couldnt in 2 months
Thank you, this explanation makes the topic much easier to understand!
How do you calculate the slope "m" in a non cartesian coordinate system ? (non ortho-regular system)
plug it into excel :P
Hi Sal! What program are you using to make those straight line "masks" (where you orient the line "mask" 1st and then draw over it) I'm not seeing this feature in Smoothdraw...I see drawing straight lines but nothing that lets you lay down the line "mask" and then draw so it only fills in inside the "mask" (e.g. 1:46 of this video)
thanks I actually have a better understanding now
Awsome Vedio. Love from Pakistan.
Thanks for the useful video. What software,platform you are using for writing is it ipad ?
First, THANKS.
Second, please resolve 2 issues:
1. Why is the line named Regression Line. From simple English language, I understand the Regression is the opposite of Progression. There should be some thought behind statisticians choosing this word to-draw-a-line-to represent-data. What's that reason?
2. For certain reason, I'm supposed to learn the Partial Least Squares Regression. Importantly, I don't have a background in Statistics. I'm finding it difficult to chart a conceptual-rung-of-ladder to achieve that goal. Can you please guide me to a series of URLs to learn PLS starting from basics through appreciating the concept nicely? This is my need and I'd be grateful if you address it.
Thanks and best regards.
Always Great Sal , Thanks
really thank you for this wonderful video!!
which software are you using to create all this?
Better than my university lecturer.
can you please share the link of the next video? i am not able to find it.
How do you do it when you start with a table of values?
Sydney Adams use the x and y from the table to plot dots on a graph
Thank you, this was very very helpful
why we are squaring the residuals ?? i case of negative values if we upply mod then it will become positive then what the case with the squares of the residuals
I have an exam in less than 10 minutes thank u 😭😭😭
Someone answer me but basically observed value minus (-) predicted value?
Thanks a lot. That was awosome
Thanks for this guide
Is there a way to watch these in the correct order? :D
I was trying to figure out the same thing
Great video...good explaination👍🏻
5:24, sorta got lost after this, can anyone help me out?
Great video! Very interesting... what software are you using to draw the lines? It seems very flexible and effective.
Pretty sure it's just Khan Academy's software.
What program is this? I am considering using it for sketching my own graphs and such
SmoothDraw 3 i just googled it, and it is free
@@maxrod98 that rhymed
@@nogapres3007 😂😂😂😂😂😂
How can get it the next video?
Great video THANKS!!!
Why are both correlation coefficient and residual "r"?? Isn't that confusing?
I still do not understand the reason why my teachers don't use these as an explanation video instead of something else.
very very clear
what is difference between residuals and bias?
fantastic explanation
Great video. Too bad I finished my stats class 2 weeks ago ]= I could of used it.
It is very useful
thank you!
Please make the board on light mode.. The writings on the board aren't visible
Thank you.
Reviewing to prepare for my interview tomorrow 😅
Nice work!
You're mixing up weight and height at around 3:30.
I'm not sure but I can see a suspicious moment in this video
Your line is approximately y = 5x - 175
if you'll give to ur model the 60 value, it returns 125. Not 150. You've drawn the line that is parallel with axis Y, but it has to be perpendicular for your regression line, doesn't it?
Who's here for finals carnegie
If the aim of the least squares is to elevate the residuals to allow them to have a larger impact in the model, why don't we elevate them by 3, 10, 1000...? That would represent an even higher impact in the model
please make tutorial videos on tensor
you forgot to mention about *unobservable* and *observable* values
I have no idea what this video is talking about...... yet.
SarcyseTiranin In which grade are you, and from which land?
you ever find out?
@@hadleyd I don't even remember leaving this comment dude. Lol.
i still don’t get it :/ you lost me at weight =m(height)+b what does that mean!! please am i like severely mentally i’ll and can’t understand rules or was it not directly explained why we suddenly wrote that like the cause not even why but the why for the why
70thviewer, 3rd liker, 3rd comment! 😁
still confused. its too fast for me
Powlowski Extension
where my precalculus gang
im here lol
Ratke Fields
I can literally NEVER understand ANYTHING from their math videos and everyone is saying how simple it is what??
Sal does good in the video, he always does, but i don't know why his voice annoys the heck out of me!
midterms 😣!
I owe you my degree. Can you be my dad ?
i somehow get more confused listening to you hop around so much. please be more organized