Isye 2027

# Suppose f1 and f0 are two known pdfs and suppose that

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: y real co and uo with 0 < uo < 1, !1 F (u) F(c) 1 u c 1 Figure 3.22: A CDF and its inverse. F −1 (uo ) ≤ co if and only if uo ≤ F (co ). Thus, if X = F −1 (U ) then FX (c) = P {F −1 (U ) ≤ c} = P {U ≤ F (c)} = F (c), so indeed F is the CDF of X. Example 3.8.12 Find a function g so that, if U is uniformly distributed over the interval [0, 1], g (U ) is exponentially distributed with parameter λ = 1. Solution: The desired exponential distribution has support R+ and CDF F given by: F (c) = 1 − e−c for c ≥ 0 and F (c) = 0 for c < 0. We’ll let g (u) = F −1 (u). Since F is strictly and continuously increasing over the support, if 0 < u < 1 then the value c of F −1 (u) is such that 110 CHAPTER 3. CONTINUOUS-TYPE RANDOM VARIABLES F (c) = u. That is, we would like 1 − e−c = u which is equivalent to e−c = 1 − u, or c = − ln(1 − u). Thus, F −1 (u) = − ln(1 − u). So we can take g (u) = − ln(1 − u) for 0 < u < 1. To double check the answer, note that if c ≥ 0, th...
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