Rotational indeterminacy n n In PCA, we choose each component in sequential fashion to account for the maximum possible amount of variation in our original data, subject to the constraint of being uncorrelated with all previously selected components. This ensures a unique solution. With the common factor model, we impose no such constraint. Therefore, there are effectively an infinite number of solutions that are identical to the extent to which they are able to approximate the matrix
Rotational indeterminacy Note: T is a orthogonal rotation matrix.
5. 2. 2 Mechanics (續)
5. 2. 2 Mechanics (續)
5. 2. 2 Mechanics (續)
5. 3 SAMPLE PROBLEM
25個屬性 Filling Family Natural Fibre Sweet Calories Plain Crisp Easy Salt Satisfying Energy Fun Kids Soggy Economical Health Regular Sugar Fruit Process Quality Treat Boring Nutritious
SMC (squared multiple correlation) n Which is the amount of variation in one variable explained by all other variables in the data set.