Prof Balka you are just the BEST! Even if I am doing multivariate statistics, I keep coming back to your channel for univariate topics as I *always* find golden nuggets in your videos. And here I am still pestering you for *more videos*... could you possibly consider making some on multivariate regression.... especially for categorical variables! My wish for Christmas. Best regards.
It's always nice to have you back, and I always very much appreciate your kind words about my videos! I've been very swamped at work for some time now, and haven't had a chance to get back to video production. I plan on getting back to it soon. (I know I keep saying that, but I mean it for reals!) I'll put it in writing now so that I have a goal -- I'll shoot for 50-100 new videos in 2018. Thanks again for the compliments.
Hi JB, Great job! I really enjoy your lectures. Just one question: At 2:29, is there a mistake in the SST formula? Is the n(I) term supposed to be there?
HI Suraj. I'm glad to hear you enjoy my video lectures! The formula for SST (treatment sum of squares) at 2:28 is correct. The n_i that appears in the formula should be there.
Thanks Kate! That's for the suggestion. Over the next year or so I'm planning on making more of my materials available (such as pdfs of some/all of the video lectures). That's been part of the plan for a while, but I just haven't found the time. I probably won't get to it until the summer or fall though. Cheers.
A question. You use d.f Error = n-k, but ive seen d.f Error = k*n-k(or cn-c) as well? I really struggle to wrap my head around the seemingly different formulas for the one way anova.
My "n" represents the total sample size (the total number of observations). If you're seeing a formula that says DF(Error) = cn-c, then they're calling the number of groups c, and letting n represent the number of observations *in each group* (in a balanced design scenario, where the number of observations is the same in each group). It would work out to the same value. Cheers.
+MMA king I've captioned a lot of my videos myself, and those are usually right on the money, but for this video they were auto generated by youtube. I'll finish captioning the rest of my videos when I get a chance.
There is nothing wrong with the formula in the video. They are (apparently) using different notation (letting n represent the number of observations within each group, not the total number of observations), and assuming a balanced design.
Professor I had you for 2040 and I miss you so much for 2050! Term test next week glad I still have these videos for reference you are the best!
Thanks Michael. I'm glad I can still be of help! Best of luck on your test!
Great job! Clearest explanation on the ANOVA table I have seen so far, keep it up!
+ha1b Thanks for the compliment!
These videos are immaculate and a lifesaver. Completely blows away any other explanations of statistics anywhere on the web. Keep up the good work!
+roux ga roux I can't ask for a nicer compliment than that! Thanks! I'm glad I could be of help.
I wish that my professor would explain concepts as clearly and concisely as you do in your videos.
Never understood why college courses are so expensive yet there’s never enough time for profs to teach the course clearly
Very clear explanation, i finally understand it now! thank you so so much!!!
Practice makes perfect...listening to more vidéos on a topic helps us understand the concepts widely and deeply..
Thanks sir
Thank you so much for such a good explanation *Sensei*
short and sweet! Thank you!
Prof Balka you are just the BEST! Even if I am doing multivariate statistics, I keep coming back to your channel for univariate topics as I *always* find golden nuggets in your videos. And here I am still pestering you for *more videos*... could you possibly consider making some on multivariate regression.... especially for categorical variables! My wish for Christmas. Best regards.
It's always nice to have you back, and I always very much appreciate your kind words about my videos! I've been very swamped at work for some time now, and haven't had a chance to get back to video production. I plan on getting back to it soon. (I know I keep saying that, but I mean it for reals!) I'll put it in writing now so that I have a goal -- I'll shoot for 50-100 new videos in 2018. Thanks again for the compliments.
You're welcome!
I hope my school can have a great stats ins like you.
Fantastic mate! - and that goes for all you videos
+Allan Gjøtterup Ernst Thanks!
haha is seriously love your videos, idk how but I've got adhd but I can actually focus on your videos and learn stuff
Hi JB,
Great job! I really enjoy your lectures. Just one question: At 2:29, is there a mistake in the SST formula? Is the n(I) term supposed to be there?
HI Suraj. I'm glad to hear you enjoy my video lectures! The formula for SST (treatment sum of squares) at 2:28 is correct. The n_i that appears in the formula should be there.
love your videos! I wish you would upload a pdf file of your lectures so I can take notes on the lecture.
Thanks Kate! That's for the suggestion. Over the next year or so I'm planning on making more of my materials available (such as pdfs of some/all of the video lectures). That's been part of the plan for a while, but I just haven't found the time. I probably won't get to it until the summer or fall though. Cheers.
ty for the simplicity
How to find the percentage contribution to the output of each factor using ANOVA?
Great! Thank you!
You are great sir
A question. You use d.f Error = n-k, but ive seen d.f Error = k*n-k(or cn-c) as well? I really struggle to wrap my head around the seemingly different formulas for the one way anova.
My "n" represents the total sample size (the total number of observations). If you're seeing a formula that says DF(Error) = cn-c, then they're calling the number of groups c, and letting n represent the number of observations *in each group* (in a balanced design scenario, where the number of observations is the same in each group). It would work out to the same value. Cheers.
Aight, thanks :)
thanks again .
So helpful.
Thanks! I'm glad I could be of help.
Thank you
thanks alot! :)
amazing... seriously..
Thanks!
Great video but compared to the other videos the subtitles are a bit messed up
+MMA king I've captioned a lot of my videos myself, and those are usually right on the money, but for this video they were auto generated by youtube. I'll finish captioning the rest of my videos when I get a chance.
i found on walpole that S^2 formula is SSE/k(n-1). a little correction for the video
There is nothing wrong with the formula in the video. They are (apparently) using different notation (letting n represent the number of observations within each group, not the total number of observations), and assuming a balanced design.
@@jbstatisticsoohh ok