Hi, I am trying to learn here! There are five TRUE/FALSE questions so I can test my understanding.

Thank you in advance!

1. With 10-fold cross-validation, 10% of the sample data are never used to estimate the model, just to

estimate the TMSE. ( training mean squared error)

2. TMSE( training mean squared error) with cross-validation is always decreasing when adding more Xs, so it is not a good estimator

for GMSE (Generalized mean squared error)

3. Using the bootstrap procedure for model averaging, every observation of the original sample will be

selected at least once.

4. One can use the AIC and BIC to test the hypothesis H0 : 5 = 0 in a logit model. (AIC is Akaike's Information Criteria and BIC is Bayesian Information Criteria )

5. The slope of the probit model is maximized when X=0.