Isye 2027

# The larger the class of estimators optimized over the

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: deviation 12 . Example 4.9.2 Suppose a fair die is rolled 1000 times. What is a rough approximation to the sum of the numbers showing, based on the law of large numbers? 158 CHAPTER 4. JOINTLY DISTRIBUTED RANDOM VARIABLES Figure 4.22: Results of simulation of sums of independent random variables, uniformly distributed on [0, 1]. (a) S vs. n. (b) Sn vs. n. n Solution: The sum is S1000 = X1 + X2 + · · · + X1000, where Xk denotes the number showing on the k th roll. Since E [Xk ] = 1+2+3+4+5+6 = 3.5, the expected value of S1000 is (1000)(3.5) = 3500. 6 By the law of large numbers, we expect the sum to be near 3500. Example 4.9.3 Suppose X1 , . . . , X100 are random variables, each with mean µ = 5 and variance σ 2 = 1. Suppose also that |Cov(Xi , Xj )| ≤ 0.1 if i = j ± 1, and Cov(Xi , Xj ) = 0 if |i − j | ≥ 2. Let S100 = X1 + · · · + X100 . (a) Show that Var(S100 ) ≤ 120. (b) Use part (a) and Chebychev’s inequality to ﬁnd an upper bound on P (| S100 − 5| ≥ 0.5). 100 4.9. LAW OF LARGE NUMBERS AND CENTRAL LIMIT THEOREM Solution:...
View Full Document

## This note was uploaded on 02/09/2014 for the course ISYE 2027 taught by Professor Zahrn during the Spring '08 term at Georgia Tech.

Ask a homework question - tutors are online