Gibbs Sampling and Data Augmentation w/ R Code | ABO Blood Typing Example

preview_player
Показать описание
Gibbs sampling with data augmentation example with R code.

Thanks for watching!! ❤️

\\Previous video (Bernstein's equations)

\\Tanner and Wong paper

\\R Code

Tip Jar 👉🏻👈🏻

♫ Eric Skiff - Chibi Ninja

#BayesianStatistics #GibbsSampler #DataAugmentation
Рекомендации по теме
Комментарии
Автор

Thanks for the explanation, this video really helped me understand data augmentation

elrohirATgmxDOTnet
Автор

Great video and example! very helpful, thanks a lot !

andreacorrado
Автор

Thanks for the video.

There is a small typo @ 2:37. The last column of AB should be 2*pa*pb.

Another question is: It seems this problem could be solved analytically. Simply plug the numbers into the probability formula. The result isn't very different.

pa * pa + 2*pa*po = 642/1583
pb * pb + 2*pb*po = 195/1583
2*pa*pb = 89/1583
po*po = 657/1583

Why add the complication of using bayesian stats?


Much appreciated if you can clarify~

taotaotan
Автор

Hi, thanks, very nice video to help me understanding how to build a joint distribution with latent variable and observation data. I suppose this the only one Vedic to combine MCMC and data augmentation technique, right? hoping you provide mere video about that, I really need it. thanks too many.

ParisZhou-zz
Автор

Thanks for the tutorial it's excellent; out of curiosity how does one get the conditional distribution to be the binomial distribution that you show? I've spent some time on this and I couldn't quite get the expression you got.

khaiyichin