If you liked this video, you can buy the corresponding ebook (plain PDF) that I used to record this video at: www.tilestats.com/ On this page, you will also find all of my videos in a logical order.
As someone who has delved into statistics, including PCA, through various sources and courses, I must say that your explanations are wonderful. I'm grateful for stumbling upon your UA-cam channel!
If you are a student: You are not alone, you cant understand everything first time you get introduced to it. Take small steps, and repeat.. over and over again. Thank you for this video, one of the best when it comes to explaining principal component analysis.
I too just stumbled across your videos. I have tutored MPH and MSc Epidemiology students in biostatistics (& epidemiology) and I know how challenging it can be to provide clear and well understood explanations. I find that your explanations are so concise and clear. I will definitely recommend them to my students and colleagues.
This is one of the best videos I've seen about PCA. Very clear and also very nice that he integrated all the information and explained some applications. Thanks a lot!
This is truly awesome! The lecture is very detailed and clear. I can only imagine the time, dedication, and hard-working that goes into creating such explanatory videos. These are very complex and abstract topics yet the presentation makes it so easy to comprehend! I am in love with this channel. Please keep on with the good works. God bless you richly
I stumbled upon your channel when I was looking simple examples (of complex algos) that can by illustrated using Excel or by hand. Your videos are great! Thank you for the effort and sharing.
@@tilestats first of all, I really love your clear explanation. I want to ask you some question, when condition we use 1/n or 1/(n-1) as our numerator of variance formula? I watch another video about calculation of covariance, although they only have 3 samples in their data, they use variance of population. in this video we have 6 samples in our data, but we use formula of variance of sample. I can't figure it out when to use what type of variance.
If you have access to the population mean use just n instead of n-1. The problem is that we usually never know the pop mean, which means that we have to estimate it from the sample and then we use n-1
It would have been easier if the illustrative examples were from the realm of quantum mechanics ... it just make sense to explain Eigenvectors and Eigenvalues when you combine them with superposition and collapse function, correct ? Why use house areas, bedrooms, and prices to illustrate ? after all, everybody understands these variables so why use them ? correct ? No, what we need to do is to incorporate quantum mechanics or biology, or the relativity theory in our illustration, so people would actually stop the video and go enroll in biology and genetics class, come back 6 months later to continue watching.
If you liked this video, you can buy the corresponding ebook (plain PDF) that I used to record this video at:
www.tilestats.com/
On this page, you will also find all of my videos in a logical order.
As someone who has delved into statistics, including PCA, through various sources and courses, I must say that your explanations are wonderful. I'm grateful for stumbling upon your UA-cam channel!
If you are a student: You are not alone, you cant understand everything first time you get introduced to it. Take small steps, and repeat.. over and over again. Thank you for this video, one of the best when it comes to explaining principal component analysis.
The best PCA video on UA-cam!
The best videos ever on the internet for use to demystify the abstractness associated with PCA. Thanks, great Tutor!
I too just stumbled across your videos. I have tutored MPH and MSc Epidemiology students in biostatistics (& epidemiology) and I know how challenging it can be to provide clear and well understood explanations. I find that your explanations are so concise and clear. I will definitely recommend them to my students and colleagues.
Thank you!
how does this video not have more views?! this was a simply brilliant intro to this topic! 👏👏
Best explanation of PCA on UA-cam. Keep it up 👍
Legend, I don't know any maths. You just made it simple
This is one of the best videos I've seen about PCA. Very clear and also very nice that he integrated all the information and explained some applications. Thanks a lot!
Excellent video for a beginner like me! thank you very much.
This is truly awesome! The lecture is very detailed and clear.
I can only imagine the time, dedication, and hard-working that goes into creating such explanatory videos. These are very complex and abstract topics yet the presentation makes it so easy to comprehend!
I am in love with this channel. Please keep on with the good works. God bless you richly
I was struggling until I found your video so thank you
Wonderful explanation in a concise way !! Thank you very mcuch sir !!
Wow!! This explanation is simply amazing. Thank you so much for the valuable content. Great work!
Thank you!
I stumbled upon your channel when I was looking simple examples (of complex algos) that can by illustrated using Excel or by hand.
Your videos are great! Thank you for the effort and sharing.
Thank you!
what an amazing video on PCA. Thank you for providing such kind of content😀
Thank you for shiring all videos.
I am so happy to find your chanel! Amazing explaination!
Your chanel deserves millions subscribers!
Thank you!
@@tilestats first of all, I really love your clear explanation. I want to ask you some question, when condition we use 1/n or 1/(n-1) as our numerator of variance formula? I watch another video about calculation of covariance, although they only have 3 samples in their data, they use variance of population. in this video we have 6 samples in our data, but we use formula of variance of sample. I can't figure it out when to use what type of variance.
If you have access to the population mean use just n instead of n-1. The problem is that we usually never know the pop mean, which means that we have to estimate it from the sample and then we use n-1
@@tilestats I got it. Thank you, Sir.
I've found your channel in very last semester from my lecturer :v wish could find your channel earlier
FANTASIC VIDEO!!!!!
Wow sir . excellent lecture .fist time commendable lecture on PCA in yutube history.what a candid and crystal idea about this topic!. U r hero sir.
Thank you!
really a clear explanation!
Simply and clear explanation
Thank you!
Thank you for the time and effort you put into these informative videos.
Would you consider doing a video on Factor Analysis?
Thank you! Yes, that is on my list.
Thank you very much sir
Thank you
A great video. I appreciate your time.
Thank you!
Best video
thank u so much
Excellent videos. I bought a lot of your pdf. You should publish more videos on Machine learning algorithms.
Thank you! Yes I will try.
please provide link or the full title of the video you are referring to as next lecture at 11:55.
Grande, fantastico. Ottimo lavoro
very useful!
this work apply on Forex Market , please confirm sir you have a example share it very I am very thankful to you.
Eigenvalue for variance covariance matrix is 12.75 and 0.389 . I was wondering how you reached to -0.8 and -0.6.
Those are the values of the first eigenvector. Watch this video to see the calculations:
ua-cam.com/video/S51bTyIwxFs/v-deo.html
TLDR; 7:04
It would have been easier if the illustrative examples were from the realm of quantum mechanics ...
it just make sense to explain Eigenvectors and Eigenvalues when you combine them with superposition and collapse function, correct ?
Why use house areas, bedrooms, and prices to illustrate ? after all, everybody understands these variables so why use them ? correct ?
No, what we need to do is to incorporate quantum mechanics or biology, or the relativity theory in our illustration, so people would actually stop the video and go enroll in biology and genetics class, come back 6 months later to continue watching.
The last example in this video is related to biology, I think.