Amen .... statistics teachers need to be taught statistics using simpler methods such as this, especially the understandability for beginners of the EXPECTED notation method. So much easier than smashing students with symbols they don't understand and then muttering about how formula A translates to formula B because of "adfjhasdfljhasbdclahbflwhef". Students all migrate directly to the bar after class, for some reason :)
I dont even know why Universities exist. Sal is providing an amazing service for free while universities provide a joke of a service for a lot of money. Sal should therefore put universities out of business but these cancerous institutions just do not disappear . This guy is what you call a teacher not some joke of a stuck up professor. No homework, no BS modules, no time wasting with commuting, just pure learning the way you want it at peace in your home; how it should be.
Yes, it is the mean but it should be differed from the mean of a sample. The expected value is the mean of a population, of what to expect as the mean of the population.
Thanks for the great video Khan! One question, so does multiple regression has the same thing? say i have b0, b1, b2 (b0 is the intercept), then do we still have b1, and b2 with the same formula? Cov(y, x1)/ var(x1), and Cov(y, x2)/var(x2)? Thank you!
"...then it would make sense that they have a negative covariance: when one goes up the other goes down, when one goes down, the other goes up." Why is something like this always said in favor of "...when one goes above its mean, the other goes below it's mean, when one goes below it's mean, the other goes above it's mean." Why does it matter if one goes up/one goes down if it's not relative to the mean? Or is this what is really meant when one says colloquially "if one goes up the other goes down"? Just trying to clarify since I'm teaching covariance soon...
Hi, its been a while since you posted but I'll try my best to answer in case anyone else has the same question. (x^2)bar is simply squaring all the x values and dividing it by the number of points. It is finding the mean of x values, however you square the x values. E.g. (2,3), (1,2) - it would be (2^2+1^2) / 2 = 5/2. (xbar)^2 is finding x-bar and literally squaring that amount. Using the prev. example - (2+1)/2 = 3/2. However this is squared so answer is (3/2)^2 = 9/4
It does not explain anything. You take the covariance definition out of nothing, do some simple algebra and state that this is a numerator in the formula for the slope. The question is: why does this formula look like so? And why covariance is defined this way?
Sal is awesome!
Einstein was right: *If one understands something properly, one can explain it simply*
المقطع نزل وعمري ٨ 🥺والحين ١٩ واحتجته لقيته بعد هذي السنين
dayum , its like pulp fiction. everything meets up in the end and story makes sense
I hope my instructor tells this instead of spreading cancer.
And I hope you dump that loser boyfriend of yours.
Amen .... statistics teachers need to be taught statistics using simpler methods such as this, especially the understandability for beginners of the EXPECTED notation method. So much easier than smashing students with symbols they don't understand and then muttering about how formula A translates to formula B because of "adfjhasdfljhasbdclahbflwhef". Students all migrate directly to the bar after class, for some reason :)
You are a Life saving god of maths!! Please do more proofs !
I just wanted to have a look into a video on regression but ended up with three days of watching dozens videos on statistics. Let me go, Khan!
KHAN IS KING!
I dont even know why Universities exist. Sal is providing an amazing service for free while universities provide a joke of a service for a lot of money. Sal should therefore put universities out of business but these cancerous institutions just do not disappear . This guy is what you call a teacher not some joke of a stuck up professor. No homework, no BS modules, no time wasting with commuting, just pure learning the way you want it at peace in your home; how it should be.
best explanation of covariance, hands down! Thank you, Sal
thanks a lot for doing the entire derivation thing, I couldn't understand two formulas for covariance while I wasn't able to derive one from another
Great video. Defiintely will be checking out many other of your statistics vids
Awesome job...Great work...Inspirational. My mind is cooking right now on lots of new ideas based on your education delivery technique here.
Yes, it is the mean but it should be differed from the mean of a sample. The expected value is the mean of a population, of what to expect as the mean of the population.
Wow. Beautifully sequenced and explained so well in a short vid..
happy new year
This video was gorgeous !
Best Teacher Ever:)
Sal,
The covariance is related to the correlation function of two variables and also convolution.
Amazing!!! thanks for the insight
Who is speaking on this? This guy is my favourite!
It's Chris Brown
Character In the video It's great, I like it a lot $$
12 Years wow still going On wow wow wow
thank you 100000000+. you are amazing!!
wow that just made soo much sense. thanks
You are the best !
I pressed Del button twice between @3:12 and @3:29 before it hit its not my screen :p
Keep up the good work, you make my day.
can you please do a video on sample co variance and why we use n-1 in the formula for sample covariance and not n-2
It helps a lot!!!!thank you!
very nicely explained.
I like this video but it could use a concrete example. Also, a covariance matrix would be a good topic
Thanks for the great video Khan! One question, so does multiple regression has the same thing? say i have b0, b1, b2 (b0 is the intercept), then do we still have b1, and b2 with the same formula? Cov(y, x1)/ var(x1), and Cov(y, x2)/var(x2)? Thank you!
so gooD!
"...then it would make sense that they have a negative covariance: when one goes up the other goes down, when one goes down, the other goes up." Why is something like this always said in favor of "...when one goes above its mean, the other goes below it's mean, when one goes below it's mean, the other goes above it's mean." Why does it matter if one goes up/one goes down if it's not relative to the mean? Or is this what is really meant when one says colloquially "if one goes up the other goes down"? Just trying to clarify since I'm teaching covariance soon...
Thank you!
can anyone tellme wheree can i get these slides khan sir teaches from in this statistics playlist? @Khan Academy
excellent!
pretty colors
What is expected value??? Is that the mean?
yes it seams
already knew it
Wondering, how old are students in Us when they are taught about covariance?
Siebe van den Elzen Around 18-19
Contradicting name ;)
?
beginning part of video helped, but getting to the middle confused me.
You forgot a closing parentheses after expected value Y
Thank you
I love you
Really complicated to follow...
whats the meaning of x times two and the other x, where is the y... explain ffs
nVariance=Sxx
I don't get the difference between x^2bar and (xbar)^2
Hi, its been a while since you posted but I'll try my best to answer in case anyone else has the same question. (x^2)bar is simply squaring all the x values and dividing it by the number of points. It is finding the mean of x values, however you square the x values. E.g. (2,3), (1,2) - it would be (2^2+1^2) / 2 = 5/2. (xbar)^2 is finding x-bar and literally squaring that amount. Using the prev. example - (2+1)/2 = 3/2. However this is squared so answer is (3/2)^2 = 9/4
how many times did he say expected value ? :D
E
keeping the colors is such a waste of effort. otherwise nice.
If facebook can worth 70billion dollar, how much would youtube worth? google got the best deal of century
facebook is a fad
It does not explain anything. You take the covariance definition out of nothing, do some simple algebra and state that this is a numerator in the formula for the slope. The question is: why does this formula look like so? And why covariance is defined this way?
Who has time for that? 15 mins drawing with different colours.....
Really grateful 🥲...always awesome
Thank you.