Information Criterion (AIC and BIC) are used to compare competing r…

Question

Information Criterion (AIC and BIC) are used to compare competing regression models. Which statement correctly differentiates AIC from BIC for model selection?

Accepted Answer

BIC penalizes model complexity more severely than AIC for large sample sizes because its penalty term grows with log(n), tending toward more parsimonious models — Here's the deal — BIC uses klog(n) while AIC uses 2k, so once log(n) beats 2, BIC hits complexity harder. You'll usually get a more parsimonious model from BIC.

Answer

AIC penalizes model complexity more severely than BIC for large sample sizes, favoring parsimonious models

Answer

AIC and BIC produce identical model rankings when the sample size exceeds 100 observations

Answer

BIC is the preferred criterion whenever prediction accuracy on fresh, unseen data is the primary goal, while AIC better suits identifying the true underlying data generating process

Information Criterion (AIC and BIC) are used to compare competing regression models. Which statement correctly differentiates AIC from BIC for model selection?

Short Explanation

Full Explanation