24. Generalized Linear Models (cont.)

MIT 18.650 Statistics for Applications, Fall 2016
View the complete course: ocw.mit.edu/18-...
Instructor: Philippe Rigollet
In this lecture, Prof. Rigollet talked about Hessian, Fisher information, weighted least squares, and iteratively reweighed least squares.
License: Creative Commons BY-NC-SA
More information at ocw.mit.edu/terms
More courses at ocw.mit.edu

[@xenosicotte]

Not many viewers left at lecture 24 !

[@pedrojaraba5436]

Estimation methods are pretty tough and it can be convenient to treat them as a black box, depending on the user :P

[@damnit258]

hhhhhhhhh survived it, it took so long,
i was taking it parallel to an NPTEL course to get a broader picture.

[@yujiaoli947]

Unique and best explanation of derivation in youtube, love this!

[@medazzouzi2649]

Heyy i have a question how to determine the g( the link function) in this case i mean if the family is not canonical how to detetmine g i m reallly stucking in this point and i need help

[@snakesnake9899]

Since 10:40 Professor changed the notation, right? I think he swaps the notation of g(.) and b(.).

[@sans8119]

Brilliant ! A masterpiece from a French man !

[@brainstormingsharing1309]

Absolutely well done and definitely keep it up!!! 👍👍👍👍👍

[@siddharthsvnit]

I wish he had covered hypothesis testing of parameter estimates !

[@vt8482]

Do you recommend any book to read about the GLM, especially on the Hypotheses testing part which is not covered in the course ? thank you

[@MOTIVAO]

I love it. There's a mistake on 10:52 Professor Rigollet said "mu is g inverse of xi transpose beta" but he wrote g prime, he should have written g inverse not prime. Just letting everyone know because it could be confusing. Great work!

[@MOTIVAO]

He notices on 17:28

[@medazzouzi2649]

Heyy i have a question how to determine the g( the link function) in this case i mean if the family is not canonical how to detetmine g i m reallly stucking in this point and i need help

[@jeffreyanderson5333]

" I need you guys, OK?" We need you too

[@siddharthsvnit]

30:25 , It only means they are uncorrelated rather than independent.

[@fastundercoverkitgoogle7381]

He is referring to normally distributed r.v. so no correlation implies independence

[@aimeokoko726]

Really useful. Thanks a lot

[@rodrigohernandezmota2770]

Finally! We start with a clean board.

[@RemcoBloemen]

That eraser though.... :|

[@zhenqin5967]

excellent course, thank you professor

[@콘충이]

Wow wow