Back to Subject

Unit 2: SUPERVISED LEARNING: REGRESSION - Practice Quiz

INT234 — Predictive Analytics 50 Questions
0 Correct 0 Wrong 50 Left
0/50

1 What is the primary goal of Simple Linear Regression?

A. To group similar data points together
B. To find the relationship between two continuous variables using a straight line
C. To classify data into discrete categories
D. To reduce the dimensionality of the dataset

2 In the linear regression equation Y = mx + c, what does 'c' represent?

A. The error term
B. The y-intercept
C. The slope of the line
D. The dependent variable

3 Which method is commonly used to estimate the coefficients in linear regression?

A. Maximum Likelihood Estimation
B. Principal Component Analysis
C. K-Means Clustering
D. Ordinary Least Squares (OLS)

4 What is a 'residual' in the context of regression?

A. The value of the independent variable
B. The slope of the regression line
C. The square of the correlation coefficient
D. The difference between the observed value and the predicted value

5 Which of the following values indicates a perfect negative correlation?

A. 0
B. 1
C. 0.5
D. -1

6 Multiple Linear Regression differs from Simple Linear Regression because it involves:

A. Multiple dependent variables
B. Polynomial terms
C. Categorical outputs
D. Multiple independent variables

7 In a regression model, if the R-squared value is 0.85, what does this mean?

A. 85% of the variance in the dependent variable is explained by the independent variables
B. 85% of the errors are zero
C. The model is 85% accurate in classification
D. The correlation coefficient is 0.85

8 Which evaluation metric is calculated by taking the average of the squared differences between predicted and actual values?

A. MAE
B. MSE
C. R-squared
D. RMSE

9 Which regression algorithm is specifically designed to predict the probability of a categorical outcome (e.g., Yes/No)?

A. Logistic Regression
B. Ridge Regression
C. Polynomial Regression
D. Simple Linear Regression

10 What function is used in Logistic Regression to map predicted values to probabilities between 0 and 1?

A. Sigmoid (Logistic) function
B. Linear function
C. ReLU function
D. Polynomial function

11 Polynomial Regression is best used when:

A. The dependent variable is categorical
B. The relationship between variables is linear
C. There are too many independent variables
D. The relationship between variables is non-linear

12 What is a potential risk of using a high-degree polynomial in Polynomial Regression?

A. Underfitting
B. Overfitting
C. Linearity
D. High bias

13 Root Mean Squared Error (RMSE) is preferred over MSE when:

A. You want the error metric to be in the same units as the target variable
B. You want to penalize outliers less
C. You want a value between 0 and 1
D. You have a classification problem

14 Which metric is less sensitive to outliers?

A. R-squared
B. RMSE
C. MSE
D. MAE

15 In OLS estimation, the assumption of 'Homoscedasticity' implies that:

A. The residuals have constant variance across all levels of the independent variable
B. The relationship is non-linear
C. The residuals follow a normal distribution
D. There is no correlation between independent variables

16 Multicollinearity in Multiple Linear Regression refers to:

A. Measurement errors in the target variable
B. The lack of a linear relationship
C. A high correlation between two or more independent variables
D. A high correlation between the dependent and independent variables

17 The range of the R-squared (R²) score is typically:

A. 0 to infinity
B. -infinity to 1
C. -1 to 1
D. 0 to 1

18 In the equation Y = b0 + b1x1 + b2x2 + ... + bn*xn, what are b1, b2, ... bn called?

A. Dependent variables
B. Residuals
C. Regression coefficients
D. Intercepts

19 Which of the following is NOT an assumption of Linear Regression?

A. Normality of residuals
B. Linearity
C. Independence of errors
D. Multicollinearity usually present

20 What is the primary advantage of Adjusted R-squared over R-squared?

A. It penalizes the addition of irrelevant independent variables
B. It can handle non-linear data
C. It is easier to calculate
D. It is always higher than R-squared

21 In Logistic Regression, the 'Logit' is defined as:

A. The natural logarithm of the odds ratio
B. The sum of squared errors
C. The square root of the variance
D. The probability of success

22 If the Pearson correlation coefficient between X and Y is 0, it implies:

A. X causes Y
B. There is a strong non-linear relationship
C. X and Y are identical
D. There is no linear relationship between X and Y

23 Which cost function is primarily used for Logistic Regression?

A. Hinge Loss
B. Log Loss (Cross-Entropy)
C. Mean Absolute Error
D. Mean Squared Error

24 When evaluating a regression model, a lower MAE indicates:

A. Better performance
B. Worse performance
C. Overfitting
D. High variance

25 To handle categorical independent variables in regression, one should usually:

A. Convert them using One-Hot Encoding (Dummy variables)
B. Use the text directly
C. Assign random numbers
D. Ignore them

26 The 'Dummy Variable Trap' occurs when:

A. Categorical variables are missing
B. One dummy variable can be predicted perfectly from the others (perfect multicollinearity)
C. There are too many categories
D. The variables are not scaled

27 Which of the following plots is best for visualizing a Simple Linear Regression?

A. Histogram
B. Pie Chart
C. Scatter Plot with a line of best fit
D. Box Plot

28 If a polynomial regression model has degree 1, it behaves like:

A. A Logistic Regression
B. A Simple Linear Regression
C. An Exponential Regression
D. A Quadratic Regression

29 In the context of MSE, what is the effect of squaring the errors?

A. It makes the calculation faster
B. It penalizes larger errors more severely than smaller errors
C. It cancels out positive and negative errors
D. It reduces the impact of outliers

30 Which algorithm minimizes the sum of squared residuals?

A. Decision Trees
B. Gradient Boosting
C. Ordinary Least Squares
D. Logistic Regression

31 What is the typical threshold used in Logistic Regression to classify a probability as '1' (or Positive)?

A. 1.0
B. 0.5
C. 0.0
D. 0.1

32 Predicting the price of a house based on size, location, and age is a problem of:

A. Classification
B. Regression
C. Clustering
D. Dimensionality Reduction

33 Predicting whether an email is 'Spam' or 'Not Spam' is a problem of:

A. K-Means
B. Linear Regression
C. Polynomial Regression
D. Classification (e.g., Logistic Regression)

34 In the equation Y = b0 + b1*X + e, what does 'e' represent?

A. The predicted value
B. The correlation
C. The error term (noise)
D. The intercept

35 If the correlation coefficient (r) is 0.9, the Coefficient of Determination (R²) is:

A. 0.9
B. 0.81
C. 0.45
D. 0.18

36 Which of the following indicates the strongest relationship?

A. Correlation = 0.1
B. Correlation = -0.8
C. Correlation = 0.5
D. Correlation = 0.7

37 Why might one perform a log transformation on the dependent variable in regression?

A. To make the distribution more normal or linearize a relationship
B. To increase the number of outliers
C. To reduce the sample size
D. To turn it into a categorical variable

38 If a model has an RMSE of 10 and an MAE of 2, what does this suggest?

A. The model is perfect
B. There are likely large outliers in the errors
C. RMSE is calculated incorrectly
D. The model is underfitting

39 In polynomial regression, if the curve passes through every single data point perfectly, the model is likely:

A. Ideally fitted
B. Overfitted
C. Underfitted
D. Linear

40 The decision boundary in Logistic Regression is:

A. Curved
B. Linear (in the feature space)
C. Undefined
D. Circular

41 What is the slope of the line y = 3x + 5?

A. 3
B. 5
C. 8
D. 0

42 When comparing two regression models on the same dataset, the one with the ___ is generally preferred.

A. Higher R-squared and Lower RMSE
B. Higher MAE
C. Lower R-squared
D. Higher RMSE

43 Which technique helps checks for 'Linearity' in a regression model?

A. Residuals vs Predicted Values plot
B. Bar chart
C. Histogram of the target
D. Correlation matrix

44 The 'Bias-Variance Tradeoff' implies that:

A. We want high bias and high variance
B. Simple models have high variance
C. As we increase model complexity, bias decreases but variance increases
D. Complex models have high bias

45 Feature scaling (normalization/standardization) is particularly important for:

A. Simple Linear Regression with one variable
B. Decision Trees
C. Ordinary Least Squares solution
D. Regression using Gradient Descent optimization

46 If R-squared is 1.0:

A. The model explains 100% of the variability in the data
B. The correlation is 0
C. The model predicts the mean for every observation
D. The model is completely wrong

47 Which of the following is a dependent variable in a study of how study time affects exam scores?

A. Subject
B. Student ID
C. Study time
D. Exam scores

48 In the context of regression, 'extrapolation' refers to:

A. Removing outliers
B. Predicting values outside the range of the training data
C. Predicting values within the range of training data
D. Calculating the mean

49 The correlation coefficient ranges between:

A. 0 and 1
B. -infinity and infinity
C. -1 and 1
D. 0 and 100

50 Which of the following is NOT a metric for Regression?

A. MAE
B. MSE
C. R-squared
D. Accuracy
Previous MCQ
Unit 1
Next MCQ
Unit 3