A data science team is building a predictive model with 50 potential predictors. They want to select a subset that generalizes well to new data. Which approach is most appropriate?

Select the model with the highest R-squared.

Use stepwise regression and rely solely on p-values to decide which predictors to include.

Use cross-validation to compare models with different subsets of predictors and select the one with the lowest cross-validated error.

Include all 50 predictors in the model because more data always improves predictions.

What does AIC (Akaike Information Criterion) balance in model selection?

The significance of each coefficient

Bias and variance of the coefficients

Goodness of fit and model complexity

The number of predictors and the sample size

A data analyst is comparing two linear regression models for the same outcome. Model A has 3 predictors and an R-squared of 0.85. Model B has 7 predictors and an R-squared of 0.87. Which model should be preferred, and why?

Model B, because it has a higher R-squared.

Model B, because more predictors always lead to better predictions.

Model A, because R-squared is the only measure of fit that matters.

Model A, because it is more parsimonious and the difference in R-squared is small.

What does the intercept term β0 represent in a simple linear regression model y = β0 + β1x?

The slope of the regression line

The predicted value of y when x equals zero

The correlation between x and y

The standard error of the estimate

Dell Data Science Foundations

Domain 4Objective 3

Linear Regression DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 3)

Part of the Advanced Analytics - Theory, Application, and Interpretation of Results for Eight Methods domain, which accounts for 40% of the DATA-SCIENCE-FOUNDATIONS exam.

32questions here

7free pages

12concepts

40%of the exam

12 Concepts in This Objective

Linear regression fundamentals
Ordinary least squares estimation
Model assumptions
Coefficient interpretation
Goodness-of-fit measures
Hypothesis testing for coefficients
Confidence intervals for predictions
Residual analysis
Multicollinearity
Categorical predictors and dummy variables
Model selection and validation
Limitations and alternatives

Questions 11–15

11application · medium
A data science team is building a predictive model with 50 potential predictors. They want to select a subset that generalizes well to new data. Which approach is most appropriate?
Select an answer first
12foundation · easy
What does AIC (Akaike Information Criterion) balance in model selection?
Select an answer first
13application · medium
A data analyst is comparing two linear regression models for the same outcome. Model A has 3 predictors and an R-squared of 0.85. Model B has 7 predictors and an R-squared of 0.87. Which model should be preferred, and why?
Select an answer first
14foundation · easy
In a regression with a categorical predictor having three categories (A, B, C), how many dummy variables are needed?
Select an answer first
15foundation · easy
What does the intercept term β0 represent in a simple linear regression model y = β0 + β1x?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to DATA-SCIENCE-FOUNDATIONS

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Dell Technologies. “DATA-SCIENCE-FOUNDATIONS” is a trademark of its owner, used for identification only.