Logistic regression: or what is the probability of success?

preview_player
Показать описание
Previous lectures:

1. Introduction to statistics

2. Linear regression vs. Statistical tests

Why do we need logistic regression
for predictions
for studying how things influence other things
Since logistic regression can easily handle both numerical and categorical predictors, it gives you the power to check literally anything on it’s ability to increase or decrease the probability (or the odds) of success.

Providing probabilities makes a logistic regression one of the most useful statistical tools for understanding the world. Such high interpretability is much more useful and pragmatic as compared to many other fancy machine learning techniques (generally called - classification). However, modern machine learning models may have a better predictive (or classification) performance, as compared to the logistic regression. Thus, if the only thing you care about is a prediction, learn other classification techniques (Random Forest or Neuronal Networks to name a few), but if you want to know how things work and what factors influence the probability of success, a logistic regression is the way to go!

Рекомендации по теме
Комментарии
Автор

That was just wholesome! Every time I see your content I end up so excited and enthusiastic, wanting to know more and more Thanks for the labor you do Dr. Zablotski! You have helped me made my mind up about studying Master and PhD degrees in Statistics. I love this ❤ !!!!

WilOspinoC
Автор

Deeply enriching and hugely informative, with all of the concepts explained in a clear and digestible fashion. Many thanks, Yury. Looking forward to your next offering

robertreid
Автор

Thank you for this awesome lecture. It is so exciting to hear it. I love to hear more about other statistics like in machine learning or AI. It was easy listening for me. Keep it up!

MySongboy
Автор

Thank you for this awesome lecture! So may I ask how can we manually adjust cutoff value for logistic regression in R?

buraktiras
Автор

plz make a lecture on ARIMA time series with R.

IAKhan-kmph
Автор

Thank you for this awesome lecture! So may I ask how can we manually adjust cutoff value for logistic regression in R?

buraktiras