Dude. This video is everything. I'm a visual learner and this one video has helped me understand this concept better than years of statistics study and multiple textbooks. Thank you!
100% agree although slight correction is that the visual, kinesthetic, etc types of learners are a myth and has been debunked. The reason we can understand it well now because the teacher is a good teacher and know how to explain the concepts intuitively. Hehe sorry for being picky, just wanna correct the misconception
You're teaching me everything I am supposed to be learning in my graduate level statistics classes. You have no idea how helpful your channel is. Thank you!
You are such a great instructor who can explain complex concept in a very crystal clear and simple way. I've followed all of your Descriptive Statistics series videos. They make me understand the statistics concepts which is crucial for my AI study much easier. Thank you!
instant subscribe and a like. great video. Edit: after completing the vid, i want to like it multiple times. In high school, i just memorized the formula for variance and all , never liked statistics as a result. but here you are, making me love it. thank you, honestly.
Dude thank you, the janky Monty Carlo book I got just threw all these definitions and formula at me in the first chapter and I had no idea what the hell a Rth moment meant but this makes sense.
I watched the video and was shocked at how well this subject was explained. There are so many parts of math out there that can be explained so easily but frustratingly are not. Thank you so much for being a light at the end of the tunnel.
Even if I'm not good at maths and all that logical physics stuff... You explained it so clearly.. that I understood the concept... Thanks a lot.. for such a helpful video.
An easy visual method to prove this statement is to observe y=x^2 graph. Starting anywhere on the graph, moving 2 points (1 left, 1 right) the same distance left and right (to keep the mean the same) on x-axis causes a larger increase in y and smaller decrease in y (works no matter you start at x>0 or x
Amazing video. Adding to it, 8:55 you need not to do Average Squared "Distance", you may easily get it by Actual(u2)-Standard(u2), which in your case is (229.8)-(225)=4.8 . Amazing Explanation. Thanks.
Please do go into the excruciating derivations of second & third moments- both sample and population! In the SAMPLE third moment for example, it makes sense algebraically that (n-1)(n-2) factors out, but since when is n-2 make sense with regards to degrees of freedom? So on with the 4th moment?
To achieve the purpose of "netting out the effect of the previous moment" come up as a subtraction of the first moment, but in subsequent moments, netting out the previous moment comes up as dividing by the standard deviation to the corresponding power?
Thanks your video is really awesome...you have makes everything crystal clear... Now , i do request to you please make a lecture video for inference topic "COMPLETENESS" and UMVUE It will be great helpful....
If there is df = n-1 for variance because of population mean, and df = n-2 for skewness because of population mean and variance, so why there is df = n-3 when you calculate kurtosis? what is the population parameter we account for?
At 3.28, you said add 12+14+17+18 and divide by 5 to get the mean, but if you do so you get the mean as 12, if you divide by 4 you get the mean as 15 , so which one is correct?
This is awesome. An actually understandable explanation. This has really helped me think through the physical meaning of moments. I'm wondering though, how you get decimal moments? In cloud physics a particle size distribution can be described with moments, and e.g. mass-weighted particle fall speed is proportional to M3.5 - but what does that actually represent physically?
As a feedback, could you please give these concepts a context. perhaps from engineering. As a bonus thing on top of the current content and structure of the video.
Because the first moment is centered on 0. Notice how all the other centered formulas have the value x MINUS the mean. That's what "centering" means: by taking away the mean, you recalibrate, or in other words "center", the formula around the mean. Intuitively, it's like shifting the whole set of axes to have the mean as your new origin point. The first moment, instead, is special because it's the formula for the mean itself, so you don't have it yet, but you still need a reference point to "center" the formula. And that reference point is 0: again, take x MINUS the reference point, which in this case is still x (x-0=x). Why 0? No reason in particular, that's just the number we chose to have as our origin point because it makes the calculations easier.
Nice Explanation! But One query, When measures like variance and Std dev are present to describe the spread, why this term "moments" are again defined? ( just to be able to compute skewness and Kurtosis ?)
I have to say that the use of distance here is probably not the best. If you include a negative number in your notion, your concept falls apart. Recall that it is exactly because of the issue we have with negative deviations that we have interference with our variance calculation, which is why we take squares or use abs value. We are more interested in the average POSITION of the data on the number line, rather than distance which would involve Abs Value function. In this case, the first moment is finding the Center of Mass to the data set. This is why everything balances out when we do Sigma(x - xbar) because we are referencing the center of mass. That's why the moments are called centered. For the Variance, we find the average area of squares given by deviations from the mean. We take those negative deviations and have them generate a positive area, then add up all of those areas. When we divide by n we are saying that each data point on average contributes s^2 to the total area. The sqrt gives the length of the leg, s, of this avg square.
![[🔴LIVE] งานประกาศรางวัล MAYA TV AWARDS 2024 : มายามหาชน](http://i.ytimg.com/vi/nwmWYXXVwBY/mqdefault.jpg)
Dude. This video is everything. I'm a visual learner and this one video has helped me understand this concept better than years of statistics study and multiple textbooks. Thank you!
100 %
100% agree although slight correction is that the visual, kinesthetic, etc types of learners are a myth and has been debunked. The reason we can understand it well now because the teacher is a good teacher and know how to explain the concepts intuitively. Hehe sorry for being picky, just wanna correct the misconception
same goes to me. it becomes difficult to understand without vizualization
I googled for days someone to explain me moments. I finally understand. Don't go to university to learn moments, come here.
i am the one who went to university and can't understand it there and understand it here! thank you Zedstatistics!
You're teaching me everything I am supposed to be learning in my graduate level statistics classes. You have no idea how helpful your channel is. Thank you!
You are such a great instructor who can explain complex concept in a very crystal clear and simple way. I've followed all of your Descriptive Statistics series videos. They make me understand the statistics concepts which is crucial for my AI study much easier. Thank you!
instant subscribe and a like. great video.
Edit: after completing the vid, i want to like it multiple times. In high school, i just memorized the formula for variance and all , never liked statistics as a result. but here you are, making me love it. thank you, honestly.
The moment these moments got clicked in my brain, that moment I understood these moments perfectly
Thank You for this wonderful video 💯👌👌
Dude thank you, the janky Monty Carlo book I got just threw all these definitions and formula at me in the first chapter and I had no idea what the hell a Rth moment meant but this makes sense.
I watched the video and was shocked at how well this subject was explained. There are so many parts of math out there that can be explained so easily but frustratingly are not. Thank you so much for being a light at the end of the tunnel.
After lots of study, watching videos, reading books finally you did better all these. Thank you
Finally a really good and clear explanation! Thank you so much!
Even if I'm not good at maths and all that logical physics stuff... You explained it so clearly.. that I understood the concept... Thanks a lot.. for such a helpful video.
You made my life easier! You really teach complicated stuff in an easy
way to understand
Even for an FRM candidate, this is the best explanation I found. Great method and excellent video ..many thanks
You sir, deserve a hug
Edit: Can you do characteristic functions next?
@6:20 - The Low Numbers dont "detract" as much as the High Numbers "inflate" this calculation - BAM !!
Double bam!
Statquest fans hi there😆😆
An easy visual method to prove this statement is to observe y=x^2 graph. Starting anywhere on the graph, moving 2 points (1 left, 1 right) the same distance left and right (to keep the mean the same) on x-axis causes a larger increase in y and smaller decrease in y (works no matter you start at x>0 or x
You are SIGMA...the way you summed up all the concept in just few minutes....Thank you!
Your video letted me know the natural meaning of some statistical concept.It really healpful.Thx!!
I have stats test today. Yesterday I barely understood the variance. Now i'm confident I'll at least not fail
wish i had a teacher like you!
you are diamond
something so fundamental made hard to understand in lecture but easy here. thanks!
Thanks to these videos, I might be able to make it through Casella & Berger, at long last!
Amazing video. Adding to it, 8:55 you need not to do Average Squared "Distance", you may easily get it by Actual(u2)-Standard(u2), which in your case is (229.8)-(225)=4.8 .
Amazing Explanation. Thanks.
good way of explanation .please also help to explain the application of higher moments
Dude, l understood very well moment. your explanation really sorted and effective 👍
Thank you ! this video is ingenious. You make an super abstract idea crystal clear
Superb Video.... Couldn't have been explained better...
What a great and helpful video! Never understood moments until that!
Please do go into the excruciating derivations of second & third moments- both sample and population! In the SAMPLE third moment for example, it makes sense algebraically that (n-1)(n-2) factors out, but since when is n-2 make sense with regards to degrees of freedom? So on with the 4th moment?
Your videos are really great ! Very well done, clarifying and fun. Thank you very much.
Amazing video mate
Absolutely marvelous job on this video. Thank you very much
Thank you very much for this video! It is very educational and well-arranged! :)
To achieve the purpose of "netting out the effect of the previous moment" come up as a subtraction of the first moment, but in subsequent moments, netting out the previous moment comes up as dividing by the standard deviation to the corresponding power?
Thanks your video is really awesome...you have makes everything crystal clear...
Now , i do request to you please make a lecture video for inference topic "COMPLETENESS" and UMVUE
It will be great helpful....
Subscribed because of this video, thank you
I finally understood this concept, all thanks to you!
Very intuitive and straightforward! Ty!
This video made my day!
Thanks!
Thank you so much !
Clearly explained !
You're welcome here anytime, JJS :)
Thank you very much. This is exactly what I needed. You are a genius.
You rock! Amazing video. I finally got a clear understanding. Thanks!!
Wowww amazing video 😍😍. This is how one must teach the statistics 😍
If there is df = n-1 for variance because of population mean, and df = n-2 for skewness because of population mean and variance, so why there is df = n-3 when you calculate kurtosis? what is the population parameter we account for?
Thank you so much! Your explanation is easy to understand.
So a population parameter you always need to know is the size of the population, right? What if you don't have that?
This is really helpful, thank you!
Thank you for these great videos ❣️
you are love in the sky of statistics ❤
Well explained! Thank you.
Would it be possible to do a video on cumulants at some point?
At 3.28, you said add 12+14+17+18 and divide by 5 to get the mean, but if you do so you get the mean as 12, if you divide by 4 you get the mean as 15 , so which one is correct?
wow sir... you are amazing
Amazing, really appreciate this
excellent explanation
Fantastic video.
Thank you sooo much for explaining this in this way
Amazing explanation! Thank you so much!
One question I hope you read it and make a video on it, why estimation causes us to loose degree of freedom?
Thank you so much for the video!! :)
so well explained
This is awesome. An actually understandable explanation. This has really helped me think through the physical meaning of moments.
I'm wondering though, how you get decimal moments? In cloud physics a particle size distribution can be described with moments, and e.g. mass-weighted particle fall speed is proportional to M3.5 - but what does that actually represent physically?
Could you please explain why Moments are generated and what do they tell about data distribution?
cyanide 4U ...and try these videos from lesson 1. !
(Regards)
th-cam.com/video/tVDdx6xUOcs/w-d-xo.html
anyone could explain why we need to use sigma (the standard deviation)
please watch his video on standard deviation. just go to his channel and you'll find loads of videos about almost everything in statistics
Because of the units. The standard deviation is on the same units as the original data.
You are my lifesaver
Can someone please tell me where is outliers video?
Fantastic. Thank you.
this is a brilliant explanation, thanks!
Thank you very much !
well explained and thanks much for your explanation
Amazing video!
Thank you very much
As a feedback, could you please give these concepts a context. perhaps from engineering. As a bonus thing on top of the current content and structure of the video.
My nigga said green at the start that shit shook me hard fam
Thank you for providing this!
that was a hell lot of help, thank you so much.
Wow! Brilliant.
Could you please make a video on the moment generating functions?
A great supplement to the BCA masters!
Of which I myself am a graduate :)
Why don't we have a centered first moment if all others are centered ?
Because the first moment is centered on 0. Notice how all the other centered formulas have the value x MINUS the mean. That's what "centering" means: by taking away the mean, you recalibrate, or in other words "center", the formula around the mean. Intuitively, it's like shifting the whole set of axes to have the mean as your new origin point. The first moment, instead, is special because it's the formula for the mean itself, so you don't have it yet, but you still need a reference point to "center" the formula. And that reference point is 0: again, take x MINUS the reference point, which in this case is still x (x-0=x). Why 0? No reason in particular, that's just the number we chose to have as our origin point because it makes the calculations easier.
Thank you !
@@utkarshbajpai5035 You're welcome! I hope it was clear enough :D
Very very helpful
Thank you, you are very clear and logical :)
Nice Explanation! But One query, When measures like variance and Std dev are present to describe the spread, why this term "moments" are again defined? ( just to be able to compute skewness and Kurtosis ?)
Thanks!
great explanation. so what do they mean when they say generalized moments? is it all four put together?
thank you !
Thanks for this video
Thanks alot... this video is amazing Bro
I have to say that the use of distance here is probably not the best. If you include a negative number in your notion, your concept falls apart. Recall that it is exactly because of the issue we have with negative deviations that we have interference with our variance calculation, which is why we take squares or use abs value. We are more interested in the average POSITION of the data on the number line, rather than distance which would involve Abs Value function. In this case, the first moment is finding the Center of Mass to the data set. This is why everything balances out when we do Sigma(x - xbar) because we are referencing the center of mass. That's why the moments are called centered. For the Variance, we find the average area of squares given by deviations from the mean. We take those negative deviations and have them generate a positive area, then add up all of those areas. When we divide by n we are saying that each data point on average contributes s^2 to the total area. The sqrt gives the length of the leg, s, of this avg square.
Beautiful.
so clear thank you so much!!!!
i was lucky to find this video :)
You are a legend!
Awesome Video ! Thanks
that's amazing
legend
Is moments can bi negative??
Love it
Uauuu super good, thak you!
I like it. Thanks
Love you zed