MSC_ASM_QUIZ4

1) Which statistic is often used to decide whether to remove a variable in backward regression?

• Shapiro Wilk

• Akaike Information Criterion (AIC)

• Durbin-Watson statistic

• Variance Inflation Factor (VIF)

2) What is the first step in backward elimination?

• Compute the Adjusted R-squared

• Select a random variable to eliminate

• Remove the least significant predictor

• Add the most significant predictor

3) What happens if no predictor has a p-value greater than the threshold in backward elimination?

• The variable with the highest p-value is removed

• The variable with the lowest p-value is removed

• The process stops, and the final model is obtained

• The process continues until only one variable remains

4) Forward selection is computationally:

• Equivalent to logistic regression

• Less efficient than backward elimination

• The same as backward elimination

• More efficient than backward elimination

5) Which function is used as the activation function in multi-logistic regression?

• ReLU

• Sigmoid

• Softmax

• Tanh

6) What is the output of a logistic regression model?

• A continuous value

• A probability between 0 and 1

• A categorical variable

• A tree-based structure

7) A telecom company wants to predict whether a customer will switch to another provider based on their call usage, monthly bill, and internet speed.

Which regression model would be the best fit?

• Linear Regression

• Polynomial Regression

• Multiple Linear Regression

• Logistic Regression

8) A biologist is studying bacterial growth and finds that the relationship between time and bacteria count follows a curved pattern rather than a straight line.

Which regression technique should be used to model the data?

• Logistic regression

• Multiple Linear regression

• Linear regression

• Polynomial Regression

9) Which issue can arise when using high-degree polynomial regression?

• Underfitting

• Overfitting

• Increased interpretability

• Decreased flexibility

10) What is the role of the threshold in logistic regression?

• It determines the probability of an event occurring

• It controls the regularization strength of the model

• It is used to classify the predicted probability into binary classes

• It reduces the correlation between independent variables