1. Regression with 1 dependent + 3 independent variables

✔ Multiple linear regression
❌ One hot encoding
❌ Simple linear regression
❌ Interaction terms

---

2. Which are categorical variables?

✔ Shirt country of manufacture
✔ Shirt size
✔ Shirt type
❌ Shirt inventory (numeric → NOT categorical)

---

3. No two independent variables are highly correlated — testing which assumption?

✔ No multicollinearity assumption
❌ No homoscedasticity
❌ No linearity
❌ No normality

---

4. Interaction terms affect changes in the _____ of the dependent variable.

✔ mean
❌ assumption
❌ category
❌ multicollinearity

---

5. Adjusted R² — correct statements:

✔ It can vary from 0 to 1.
✔ It penalizes unnecessary explanatory variables.
✔ It is a regression evaluation metric.
❌ It is greater than 1.

---

6. Stepwise method starting with full model?

✔ Backward elimination
❌ Extra-sum-of-squares F-test
❌ Forward selection
❌ Overfit selection

---

7. Performs well on training but poorly on unseen data = too much?

✔ Variance
❌ Bias
❌ Entropy
❌ Leakage

---

8. Which regularization removes variables completely?

✔ Lasso regression
❌ Elastic net regression
❌ Ridge regression
❌ Independent regression

---

9. Suitable for multiple linear regression?

✔ Utility company predicting electricity usage (many predictors)
✔ Marketing team predicting sales from multiple inputs
❌ Educational researcher (only 1 predictor → simple regression)
❌ Bank fraud detection (classification, not regression)

---

10. Higher residuals for women → violates which assumption?

✔ Homoscedasticity (unequal variance across groups)
❌ Linearity
❌ Independent observations
❌ No multicollinearity

---

🧾 Summary Table of All Answers

---