From the course: Machine Learning and AI Foundations: Classification Modeling
Unlock the full course today
Join today to access over 24,800 courses taught by industry experts.
Logistic regression
From the course: Machine Learning and AI Foundations: Classification Modeling
Logistic regression
- Okay, let's talk about logistic regression. You're gonna notice some similarities in look and feel from logistic regression and discriminate analysis, particularly at the level of detail, but once we get to the other algorithms, you're gonna notice a striking difference between logistic and discriminate on the one hand, and all of the others, because these are really the two that come from the statistics tradition. First, note that like many of the algorithms, virtually all, in fact, logistic operates by listwise deletion, meaning that if any data's missing, that row is eliminated. All inputs are used unless you choose to do stepwise logistic. So, like discriminate, we have a stepwise option. Have to use scale variables. In practice, however, when epidemiologists, health researchers, use logistic regression, they tend to have mostly categorical variables, so we can employ dummy coding, and dummy coding is very common in logistic regression models. Logistic is arguably the most…
Contents
-
-
-
-
-
Overview2m 10s
-
(Locked)
Discriminant with three categories5m 44s
-
(Locked)
Discriminant with two categories5m 2s
-
(Locked)
Stepwise discriminant1m 3s
-
(Locked)
Logistic regression10m 54s
-
(Locked)
Stepwise logistic regression1m 3s
-
(Locked)
Decision Trees4m 46s
-
(Locked)
KNN3m 58s
-
(Locked)
Linear SVM8m 2s
-
Neural nets7m 57s
-
(Locked)
Bayesian networks7m 54s
-
(Locked)
Heterogenous ensembles3m 22s
-
(Locked)
Bagging and random forest3m 26s
-
(Locked)
Boosting and XGBoost1m 57s
-
-
-