In the following plot you see the rejection frequencies (defined as the frequency in which the estimator is non-zero) for 100 regressors (represented along the horizontal axis), and repeated for the...

Extracted text: In the following plot you see the rejection frequencies (defined as the frequency in which the estimator is non-zero) for 100 regressors (represented along the horizontal axis), and repeated for the OLS, Lasso and Adaptive Lasso estimators. The red dots are the estimates for the true non-zero coefficients. The black dots are the estimates of true-zero coefficients. OLS Lasso Adaptive Lasso 12- 12- 1.2- 0.8- 0.8- 0.8- 0.4- 0.4- 0.4- 0.0- 0.0- 0.0- 100 100 You can learn from the plots that: O a. The Lasso has a clear shrinkage towards zero O b. The Adaptive Lasso has better model-selection properties because it estimates more frequently the true zeros as zeros O. All of the options Od. The OLS has poor predictive performance O e. The Adaptive Lasso has better model-selection properties because it estimates more frequently the true non-zeros as non-zeros coef *8-