R - Binary Logistic Regression Lecture & Example

preview_player
Показать описание
Lecturer: Dr. Erin M. Buchanan
Missouri State University
Spring 2016

This video covers how to run and interpret a binary logistic regression using glm - from assessing ratio of cases, additivity, to understanding the output.

Рекомендации по теме
Комментарии
Автор

Thanks Dr. Erin M. Buchanan. I'm your fan, from Colombia.

ginomarttelocarmonadiaz
Автор

I have watched every single one of your tutorials and consider myself a master at all of them... except this one. Still have no clue what just happened :D

brittjane
Автор

Min 26:30 - A super quick way of doing percents is using "prop.table" instead of "table" (just discovered this recently): percentage <- * 100

adionicas
Автор

BayorEdPsych isn't on CRAN anymore, but library(fmsb) has an easy function for r-squared: NagelkerkeR2(model)

adionicas
Автор

Hi Erin Thanks a lot for these tutorials. They are really good and help a lot. I need to run a logistic model with an interaction term, is there any limitation for logistic models to include interactions? In my case the interaction is between a continuous variable and another binary variable. I had no problem running it. But since you didn't include an interaction term I was wondering if there was some limitation.
Thanks again. Although there are many statistic channels I always come to yours first.

aplysiaenelpaisdelasmaravi
Автор

Thank you so much for your videos; I find your explanations very helpful :) What would be a function to estimate the effect size for a mixed effect logistic model (glmer)?

olgaprocenko
Автор

I have a quation, please answer me. Is this method for logistic regression as statistical model or machine learning? because I need just code in R for normal (statistical logistic regression) for my work.

ariIran
Автор

Is there any way to access your notes and data for this video? It's a wonderful video, and a great walk through of the process!

josaint
Автор

@19:20, install the R package BaylorEdPsych: Warning in install.packages :
package ‘BaylorEdPsych’ is not available (for R version 4.0.0)

jarrelldunson
Автор

@16:38, when iI run the model, I get: "Error in eval(family$initialize) : y values must be 0 <= y <= 1" error. I convert to Factors...
master$CHILDREN <- as.factor(master$CHILDREN)
master$RACE <- as.factor(master$RACE)
master$WORKSTAT <- as.factor(master$WORKSTAT)

jarrelldunson
Автор

talks super fast and flits between pages many times. cannot follow what is happening

jababnamgay