Which of the following statements correctly describes the difference between linear discriminant analysis LDA and quadratic discriminant analysis QDA? Choose all that apply.
1 - LDA estimates the class priors as the proportion of examples of each class.
2 - QDA estimates the same priors assuming a Bernoulli distribution where p=0.5. Compared to LDA, QDA tends to have higher bias but lower variance.
3 - LDA assumes the same covariance matrix per class. QDA assumes a different covariance matrix per class.
4 - Both LDA and QDA assume the distribution for every class is Gaussian.