An introduction to the concept of a sufficient statistic

[@tylerforrester11]

Thank you, Ben. This is a wonderfully clear explanation of sufficient statistics.

[@innocentabaa2979]

Thanks so much, Ben. My Bayesian data analysis course is becoming simpler all thanks to your videos

[@rvolkov1]

Thanks so much for this explanation. My professor did not explain any reasoning or intuition regarding the calculations and just began showing examples of calculations. You showed in ~6 minutes what he could not in >25 minutes of lecture.

[@vsenderov]

This video sufficiently well explained the concept.

[@jivillain]

OMG you are 300% better than my professor of inference on conveying a concept

[@harveycheng2085]

Actually 500% than mine

[@harveyg9898]

Hi there, thanks a lot for the video! I think that the sign in front of (N-t) should be a + not a -

[@SpartacanUsuals]

Yes, you’re right. Thanks for pointing this out. I suspect I’ll leave the video as is (and perhaps add a note) since this mistake isn’t central to understanding the concept. Thanks again! Best, Ben

[@YairCat]

Noticed that as well, well done

[@jingzhan7868]

At 5:34, Log L = tlog(θ) + (N-t)log(1-θ), it should be addition, not aubtraction!

[@selinashitongli6838]

Upvote, please fix!!

[@walkerzhong1279]

Thanks for your well-organized explanation!!

[@siko993]

Well that was a sufficient video, thanks!

[@PedroRibeiro-zs5go]

Thanks so much! That was a good explanation about sufficiency!

[@aaaaaawda]

Good and logical explained!!! Just like a good prof has to do!!!

[@andetstk1237]

This was a sufficient explanation ;) Seriously though, great video

[@monicaah2468]

which other book can we refer for this topic?

[@notgoodatmathmmm6185]

very good explanation, ty

[@ben5380]

Thank you Ben! Very helpful

[@jacoboribilik3253]

Ben, when you write down P(theta | t(X)) you are implictly giving away that you are Bayesian? Isn't the parameter a fixed unknown constant? I am asking this because in most textbooks one is usally told that a statistic T is sufficient if and only if the conditional distribtuion of the sample given t is independent of theta, but your way of approaching the issue is much more intuitive, setting P(theta | t(x)) = P(theta | X)) then t(X) is sufficient for theta.

[@jeanmichelsarr6040]

Yet, the transition from the frequentist definition to this one is not straightforward. I don't know how to be sure there is an equivalence between each definition.

[@whozz]

This video is sufficient because If I watch any other video, I won't learn anything new

[@whozz]

Just kidding :P

[@hugob8180]

Thank you for the video ! ( ps: you substracted the logs instead of adding them together)

[@wiesjevandenheerik2790]

Thanks for your video! How do you get to the logL? Where does t come from all of a sudden?

[@jasonzavaglia]

Awesome - thanks Ben

[@155_shraddhabagwe4]

how did u get the differentiation 0?

[@hongduosun9711]

Great explanation! thanks!

[@lac2275]

For example what if you didn't know the MLE of the joint is the sample mean, how would you use the sufficient statistic to help you estimate theta?.....because that is the main point....the main point is that sufficient statistic are suppose to help you make estimation of an unknown parameter when MLE or Bayes Estimation fail....in the cases when MLE or Bayes Estimation work then it is pointless to use a sufficient statistic, like in your coin flip example.

[@Prancer4799]

Thank you! this really helped

[@lac2275]

my problem is that you are using an example where MLE would get you a unique reliable estimator....sufficient statistics should be presented in cases where MLE or Bayes Estimation don't work or are cumbersome to generate.

[@jackbryde5606]

Thank you!

[@kunliu4349]

should be logL = tlogθ + (N-t)log(1-θ)

[@GeoViktorq]

Damn, bro. I love you.

[@dalkeiththomas9352]

Thank youu

[@Dilz13]

thanks

[@jonathanspurgeon5461]

a sufficient explanation

[@tedwoodsides4688]

thank u sir

[@drew96]

blue print on blue background. Am I the only one who is having trouble reading this?

[@kipngenopozee]

Blue writings invisible