Question

Answer below statements. Please also include a sentence with explanation.

Question 1: The penalty constant λ in penalized regression controls the trade-off between lack of fit and model complexity. TRUE or FALSE

.

Question 2: Elastic net regression uses penalties from both the ridge and lasso regression and hence combines the benefits of both. TRUE or FALSE

.

Question 3: If our predicting variables have high multicollinearity, we should prefer LASSO over ridge regression for model selection. TRUE or FALSE

.

Question 4: Variable selection is a simple and solved statistical problem since we can implement it using the R statistical software. TRUE or FALSE

Answer 1

1.) TRUE. As the penalty constant increases, the variance decreases and the bias increases.

2.) TRUE. Elastic net regression is the generalization method, that linearly combines the ridge and LASSO regression.

3.) TRUE. LASSO works well in case of high multicollinearity. In LASSO, one of the correlated predictor has a larger coefficient.

4.) TRUE. In R statistical software, all these variable selection and statistical problems are simple and easy to operate.