Quiz #2
How to choose the number k of principal components or partial least
squares (PLS)features in principal component regression (PCR) or PLS
Regression?
Correct!
The choice of k depends on the application and can be chosen by the
cross validation
The choice of k should be as large as possible, say, k=p.
The choice of k should be a small number such as k=2,3 or 5.
Question 2 pts
In the linear regression model , the
Laplace Prior on yields the estimator of the form
Correct! LASSO
Ridge regression
None of the above
Question 3 pts
What is the main difference between linear discriminant analysis (LDA)
and Logistic regression?
Correct!
Logistic regression is often more robust than LDA, since it does not
make any assumption on the marginal distribution of X.
How to choose the number k of principal components or partial least
squares (PLS)features in principal component regression (PCR) or PLS
Regression?
Correct!
The choice of k depends on the application and can be chosen by the
cross validation
The choice of k should be as large as possible, say, k=p.
The choice of k should be a small number such as k=2,3 or 5.
Question 2 pts
In the linear regression model , the
Laplace Prior on yields the estimator of the form
Correct! LASSO
Ridge regression
None of the above
Question 3 pts
What is the main difference between linear discriminant analysis (LDA)
and Logistic regression?
Correct!
Logistic regression is often more robust than LDA, since it does not
make any assumption on the marginal distribution of X.
