Logistic Regression with Variable Selection and Categorical Data Analysis in R

☕If you would like to support, consider buying me a coffee ☕: buymeacoffee.c...
For one-on-one tutoring/consultation services: guide-tree-sta...
I offer one-on-one tutoring/consultation services for many topics related statistics/machine learning. You can also email me [e-mail]
For rcode and dataset: gist.github.co...
Correction for mistake made on 23:45. I stated that "For every 10 year older the odds of death increases by 43% while controlling for all other predictors in the model". That statement is incorrect because I multiplied the percent change in Heart failure by one unit increase in age of 4.3 by 10 to get 43%. This was the incorrect calculation used, 10 * [(exp(.042)-1)*100]. I needed to instead multiply the log odds coefficient by 10 prior to exponentiating. This is the correct calculation (exp(.042*10)-1)*100 = 53%. Therefore, for every 10 year older the odds of death increases by 53% while controlling for all other predictors in the model. Thank you Utku Pamuksuz for spotting that.
This video is a tutorial of how to conduct some categorical analyses using R studio.
At 3:37 of the video, I meant to say -- "Though it was not, lets just say it was *dependent (i.e., significant)". Also, the pronunciation of Creatinine was a bit off, it is krē-ˈa-tə-ˌnēn.
The analyses reviewed are chi-squared test of independence, Fisher's exact test, and logistic regression. Effect size with Cramer's V for Chi-squared test of independence is covered. In addition, variable selection (i.e., model shrinkage) with stepwise regression, bootstrap, and multicollinearity detection with Variance Inflation Factor (VIF) for logistic regression models is also covered. The dataset used is from kaggle and contains patients with heart failure.