Multiple logistic regression

Intermediate Regression in R

Richie Cotton

Data Evangelist at DataCamp

Bank churn dataset

has_churned	time_since_first_purchase	time_since_last_purchase
0	0.3993247	-0.5158691
1	-0.4297957	0.6780654
0	3.7383122	0.4082544
0	0.6032289	-0.6990435
...	...	...
response	length of relationship	recency of activity

¹ https://www.rdocumentation.org/packages/bayesQR/topics/Churn

glm()

glm(response ~ explanatory, data = dataset, family = binomial)

glm(response ~ explanatory1 + explanatory2, data = dataset, family = binomial)

glm(response ~ explanatory1 * explanatory2, data = dataset, family = binomial)

Prediction flow

explanatory_data <- expand_grid(
  explanatory1 = some_values,
  explanatory2 = some_values
)
prediction_data <- explanatory_data %>% 
  mutate(
    has_churned = predict(mdl, explanatory_data, type = "response")
  )

The four outcomes

	actual false	actual true
predicted false	correct	false negative
predicted true	false positive	correct

¹ https://campus.datacamp.com/courses/introduction-to-regression-in-r/simple-logistic-regression?ex=10

Confusion matrix

actual_response <- dataset$response
predicted_response <- round(fitted(mdl))

outcomes <- table(predicted_response, actual_response)

confusion <- conf_mat(outcomes)

autoplot(confusion)

summary(confusion, event_level = "second")

Visualization

Use faceting for categorical variables.
For 2 numeric explanatory variables, use color for response.
Give responses below 0.5 one color; responses above 0.5 another color.

scale_color_gradient2(midpoint = 0.5)

Let's practice!

Intermediate Regression in R