Log odds interpretation of logistic regression

[@minhha8981]

A 5 mins lecture >>>>>> An 1 hour and a half in class lecture. Mind blowing. Thank you.

[@pedrocolangelo5844]

Ben Lambert certainly has one of the best KZbin channels ever

[@prithviprakash1110]

Great explanation there. Can't believe this was posted way back in 2013. Still relevant nearly 10 years later!

[@quocanhhbui8271]

Omg just the right explanation im looking for. My professor introduced everything out of thin air and it was hard to put pieces together even though I have some backgrounds on logistic regression.

[@jorgepriegomoreno3224]

Can't stress enough how gratefoul I'm woth you kind Sir, my college degree in economics will be dedicated to you

[@zgbjnnw9306]

Excellent explanation! Always come to your channel when I got confused in class! Thank you so much : )

[@flashguy234]

Cheers Ben, this cleared up my understanding wonderfully!

[@Azam_Pakistan]

Very nice. Please raise the audio volume, its very difficult to follow on my laptop.

[@FlyingSavannahs]

Thanks simple videos to understand. Helps with many basic concepts....sometimes requiring review. hope to see more. Taking Categorical Data Analysis in the Spring....Doctoral level and I am not a Biostats major in Public Health....I am Epi. Looking forward to watching all to help review basic concepts I may have let go of.

[@SpartacanUsuals]

Hi, many thanks for your message and kind words. If you are willing it would be great to see your syllabus; this would allow me to add these videos to the list. Best, Ben

[@vimalkumar2992]

Your video is 'To the point'

[@anooshiravanrazmavar9814]

Thank you very much professor Lambert that was very helpful.

[@rodrigomorales5841]

So to make sure; one can say that log of the odds is equivalent to the dot product (wT dot X) which is where we get our linear combination?

[@kylevietti8700]

Thanks for showing but when doing 1-p. When you say the num and denim simplifying to one (you circled in purple) isn’t it just the number values of e^betas positive and negative values equaling zero then leaving you with 1/1+e^betas.

[@zgbjnnw9306]

Is the logistic regression part of GLM? Do you have videos explaining other GLM models in which the dependent variable follows other non-normal distributions, such as Gamma, poisson, negative binomial? Thank you!

[@NONAME_G_R_I_D_]

kzbin.info/www/bejne/gGXdf6B9ptmnjNE

[@RahamadhBasha]

How to explain about linear regression and linear regression in interviewer frequently asking this questions could you make the video

[@captainalan]

this was very helpful; thanks!

[@FlyingSavannahs]

Also, are you ever available for questions? Answered by video of course. In the U.S.. we have language barriers unless one is of Asian or Indian descent.

[@SpartacanUsuals]

Hi, yes I am. I am also looking into getting my videos transcribed to help people with non-native English to understand the videos better. Many thanks, Ben

[@yadavsanderamniwas]

this is not correct definition of odds ratio. p/1-p is odds, not odds ratio. odds ratio is altogether a different thing than odds. When you have a categorical predictor variable X and target variable Y, odds ratio is the ratio of odds . odds ratio=(odds of Y when X=1)/ (odds of Y when X=0) Please research and correct me if I am wrong.

[@lastua8562]

I am not sure I understand this correctly, specifically why your X is binary which is not necessary in Logit. Could you please review your comment and elaborate, if accurate?

[@kenjiyamazaki8802]

Is this about log of odds, not log of odds ratio? I believe odds is different from odds ratio.

[@MAHANTING]

they are same i think

[@woodpeckersonfirs]

No. Logit is log odds. Odds ratio is a ratio of odds

[@kumaraswamy7848]

how can we find beta0,beta1 values

[@selfmadev85]

Using maximum likelihood

[@YouUndeground]

Thank you!

[@smakarevich]

Nice!

[@bathanthegazy6083]

Thank you

[@TheJourneyFootballVideos]

Thanks :)

[@Passang1423]

👍