Linear Regression DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 5)

Questions 21–25

1. 21expert · hard

   A data scientist is modeling the relationship between advertising spend and sales. The residual plot shows a clear curved pattern, and the Q-Q plot of residuals is approximately normal. The data scientist is considering three actions: (1) add a quadratic term for advertising spend, (2) use a log transformation of the dependent variable, (3) use a robust regression method. Which action is most appropriate?

   • ARemove the outliers that are causing the curvature.
   • BAdd a quadratic term for advertising spend to capture the curvature.
   • CUse a log transformation of the dependent variable to address the curvature.
   • DUse a robust regression method to handle outliers.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   What does R-squared measure in a linear regression model?

   • AThe proportion of variance in the dependent variable explained by the independent variables
   • BThe significance of the overall model
   • CThe correlation between the dependent and independent variables
   • DThe magnitude of the regression coefficients
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A data scientist fits a linear regression model with three predictors. The output shows that the p-value for the overall F-test is 0.001, but the p-values for two of the three individual t-tests are greater than 0.05. What is the most appropriate conclusion?

   • AThe two predictors with high p-values should be removed immediately because they are not contributing to the model.
   • BThe model as a whole is statistically significant, but not all predictors are individually significant.
   • CThe model is not statistically significant because the individual t-tests are not all significant.
   • DThe F-test is unreliable because the t-tests are not all significant.
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   What is the primary purpose of linear regression in data science?

   • ATo classify observations into discrete categories based on predictor values
   • BTo model the relationship between a dependent variable and one or more independent variables
   • CTo reduce the dimensionality of a dataset while preserving variance
   • DTo group similar data points based on feature similarity
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   In the linear regression equation y = β0 + β1x + ε, what does the term ε represent?

   • AThe error term capturing the difference between observed and predicted values
   • BThe slope of the regression line
   • CThe predicted value of y for a given x
   • DThe intercept of the regression line
   Reveal Solution Discussion Select an answer first