FREE Master of Data Science: Statistical Inference & Regression Models Questions and Answers Flashcards
6 cards from real MS-DS Master of Data science practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 6 FREE Master of Data Science: Statistical Inference & Regression Models Questions and Answers flashcards as text
In ridge regression, what effect does increasing the regularization parameter lambda have on the coefficient estimates?
Answer: Coefficients are shrunk toward zero but remain nonzero
Increasing lambda in ridge regression penalizes large coefficients, shrinking them toward zero without setting any exactly to zero.
What is the interpretation of a 95% prediction interval for a new observation in linear regression?
Answer: A new individual observation will fall within this interval with approximately 95% probability
A 95% prediction interval accounts for both estimation uncertainty and individual variation, capturing where a single new observation is expected to fall.
Which method is most appropriate for handling non-constant variance (heteroscedasticity) in a regression model?
Answer: Using weighted least squares estimation
Weighted least squares assigns different weights to observations based on the inverse of their variance, directly addressing heteroscedasticity.
In the context of hypothesis testing for regression coefficients, what does a Type II error represent?
Answer: Failing to reject the null hypothesis when it is false
A Type II error occurs when we fail to reject a false null hypothesis, meaning we miss a real effect that exists in the population.
What is the key advantage of using adjusted R-squared over R-squared when comparing regression models?
Answer: It penalizes the addition of predictors that do not improve the model
Adjusted R-squared penalizes for adding predictors that do not meaningfully improve model fit, preventing artificial inflation of the goodness-of-fit measure.
When bootstrapping is used for inference on a regression coefficient, what is the primary advantage over traditional parametric methods?
Answer: It does not require assumptions about the distribution of errors
Bootstrapping constructs confidence intervals by resampling the data, making it robust when parametric assumptions like normality of errors are violated.