Inferential Statistics PCAD-31 Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   What is multicollinearity in regression?

   • AHigh correlation between predictor variables, which can inflate standard errors and make coefficient estimates unstable
   • BThe outcome variable being correlated with the predictors
   • CThe model having too many predictors relative to the sample size
   • DThe residuals being correlated with each other
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A data scientist fits a linear regression model with 5 predictors and obtains an R-squared of 0.85 and an adjusted R-squared of 0.60. Which conclusion is most appropriate?

   • AThe model is overfitting because adjusted R-squared is lower than R-squared
   • BThe model fits well and all predictors are likely significant
   • CThe model is underfitting because adjusted R-squared is too low
   • DThe large difference between R-squared and adjusted R-squared suggests that some predictors may not be adding value
   Reveal Solution Discussion Select an answer first
4. 3application · medium

   A marketing analyst wants to model the relationship between advertising spend (in dollars) and monthly sales revenue (in dollars). The analyst suspects the relationship is approximately linear. Which modeling approach is most appropriate?

   • ALinear regression with sales revenue as the outcome and advertising spend as the predictor
   • BLogistic regression with sales revenue as the outcome and advertising spend as the predictor
   • CLinear regression with advertising spend as the outcome and sales revenue as the predictor
   • DBootstrap the data and fit a linear regression to each bootstrap sample, then average the coefficients
   Reveal Solution Discussion Select an answer first
5. 4foundation · easy

   What is a key limitation of bootstrapping?

   • AIt always produces narrower confidence intervals than traditional methods
   • BIt may not perform well when the original sample is not representative of the population
   • CIt requires the population to be normally distributed
   • DIt can only be used for the sample mean
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   Which of the following is an assumption of linear regression?

   • AThe predictor variables must be categorical
   • BThe outcome variable must be binary
   • CThere must be no relationship between the predictors and the outcome
   • DThe residuals are normally distributed
   Reveal Solution Discussion Select an answer first