This preview shows page 1. Sign up to view the full content.
Unformatted text preview: . The confdence intervals For the frst example are x 1 [ 1 . 0228 , 1 . 0017] , x 2 [1 . 0018 , 1 . 0022] and For the second example are x 1 [0 . 965 , 1 . 035] , x 2 [ 1 . 035 , . 965] . Those For the second example are 20 times larger than For the frst, since they are related to the size oF A 1 , but in both cases about 95% oF the samples lie within the intervals, as expected. Remember that these intervals should be calculated using a Cholesky decom-position or the backslash operator. Using inv or raising a matrix to the 1 power is slower when n is large, and generally is less accurate, as discussed in Chapter 5....
View Full Document
- Fall '11