Statistics Regression & Correlation Analysis 3

Question 1

What is logistic regression used for?

Accepted Answer

Predicting a binary or categorical outcome variable

Answer

Logistic regression models the probability of a binary outcome (e.g., yes/no, success/failure) as a function of predictors.

Question 2

What does the adjusted R² penalize for compared to R²?

Accepted Answer

Adding predictors that don't improve the model

Answer

Adjusted R² decreases when a predictor is added that doesn't sufficiently improve model fit, preventing overfitting.

Question 3

Which regression diagnostic plot is used to check the linearity and homoscedasticity assumptions?

Accepted Answer

Residuals vs. fitted values plot

Answer

Plotting residuals against fitted values reveals non-linear patterns and changing variance (heteroscedasticity).

Question 4

What is the purpose of polynomial regression?

Accepted Answer

To model a curved (non-linear) relationship between X and Y

Answer

Polynomial regression adds squared or higher-order terms of X to model non-linear relationships within a linear framework.

Question 5

In the context of regression, what is overfitting?

Accepted Answer

A model that fits the training data too closely and performs poorly on new data

Answer

Overfitting occurs when a model captures noise rather than the true signal, leading to poor generalization.

Statistics Practice Test