1. Learn
    2. /
  2. Courses
    3. /
  3. Introduction to Regression with statsmodels in Python
  • 1

    Simple Linear Regression Modeling

    Free

    You’ll learn the basics of this popular statistical model, what regression is, and how linear and logistic regressions differ. You’ll then learn how to fit simple linear regression models with numeric and categorical explanatory variables, and how to describe the relationship between the response and explanatory variables using model coefficients.

  • 2

    Predictions and model objects

    In this chapter, you’ll discover how to use linear regression models to make predictions on Taiwanese house prices and Facebook advert clicks. You’ll also grow your regression skills as you get hands-on with model objects, understand the concept of "regression to the mean", and learn how to transform variables in a dataset.

  • 3

    Assessing model fit

    In this chapter, you’ll learn how to ask questions of your model to assess fit. You’ll learn how to quantify how well a linear regression model fits, diagnose model problems using visualizations, and understand each observation's leverage and influence to create the model.

  • 4

    Simple Logistic Regression Modeling

    Learn to fit logistic regression models. Using real-world data, you’ll predict the likelihood of a customer closing their bank account as probabilities of success and odds ratios, and quantify model performance using confusion matrices.

Initializing

Exercise

Measuring logistic model performance

As you know by now, several metrics exist for measuring the performance of a logistic regression model. In this last exercise, you'll manually calculate accuracy, sensitivity, and specificity. Recall the following definitions:

Accuracy is the proportion of predictions that are correct. $$ \text{accuracy} = \frac{TN + TP}{TN + FN + FP + TP} $$

Sensitivity is the proportion of true observations that are correctly predicted by the model as being true. $$ \text{sensitivity} = \frac{TP}{TP + FN} $$

Specificity is the proportion of false observations that are correctly predicted by the model as being false. $$ \text{specificity} = \frac{TN}{TN + FP} $$

churn, mdl_churn_vs_relationship, and conf_matrix are available.

Instructions

100 XP
  • Extract the number of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) from conf_matrix.
  • Calculate the accuracy of the model.
  • Calculate the sensitivity of the model.
  • Calculate the specificity of the model.