Deriving the Binomial canonical link function, logit, for Generalized Linear Model (GLM)

[@cypherecon5989]

Hands down the best playlist on GLMs on youtube. Very similar to what is teached in uni lectures but better to understand.

[@Stats4Everyone]

Thank you for this comment. I hope to add more to this playlist soon. In the meantime, I'm glad that you found this video to be helpful

[@umangsaini8559]

Great explanation! Tons of help in Actuarial Sciences❤

[@TheMed177]

😍You have done an outstanding job in your explanation of the exponential family, making it remarkably clear and easy to comprehend.

[@Stats4Everyone]

Thanks! I am happy to hear that this video was helpful :-)

[@PabloRengel-bj5id]

Thank you so much. I'm from Mexico, and right now I'm studying for my master's degree. These topics are really difficult for me. I'm studing with the Adrian Dobson's book (An Introduction to Generalized Linear Models) and I don't understand anything in this book but your videos are quite clear.

[@taotaotan5671]

I am super glad that you have posted this series of videos! Thanks a ton!!
For this binomial example, are we treating the "n" as a known parameter (therefore, being ignored)? Would it be possible to factor "n" into theta?

[@Stats4Everyone]

Thanks for this thoughtful message. I wonder, in your example, would the Poisson distribution be better? In the binomial distribution we make three assumptions, 1. independent trials, 2. probability of success (pi) is constant, 3. the number of trails (n) is fixed.

[@paultrow7108]

Every time you say "K", I can't help thinking of the principal, Mr. Mackey, in South Park. "Mkay?"