Dividing this by h and letting h 0 yields 420 47 joint

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

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ∞ The integral in (4.15) can be viewed as the integral of fX,Y over the line u + v = c shown in Figure 4.14. This is an integral form of the law of total probability, because in order for X + Y = c, it is necessary that there is some value u such that X = u and Y = c − u. The integral in (4.15) integrates (and integration is a type of summation) over all possible values of u. If X and Y are independent, so that fX,Y (u, v ) = fX (u)fY (v ), then (4.15) becomes ∞ fX (u)fY (c − u)du. fS (c) = (4.16) −∞ which, by the definition of the convolution operation “∗”, means fS = fX ∗ fY (if S = X + Y , where X , Y are independent). Note the strong similarity between (4.13) and (4.14), derived for sums of integer-valued random variables, and (4.15) and (4.16), derived for sums of continuous-type random variables. 4.5. DISTRIBUTION OF SUMS OF RANDOM VARIABLES 139 Example 4.5.4 Suppose X and Y are independent, with each being uniformly distributed over the interval [0, 1]. Find the pdf of S...
View Full Document

Ask a homework question - tutors are online