13.07.2015 Views

Mathematical Methods for Physics and Engineering - Matematica.NET

Mathematical Methods for Physics and Engineering - Matematica.NET

Mathematical Methods for Physics and Engineering - Matematica.NET

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

30.7 GENERATING FUNCTIONSThe MGF will exist <strong>for</strong> all values of t provided that X is bounded <strong>and</strong> alwaysexists at the point t =0whereM(0) = E(1) = 1.It will be apparent that the PGF <strong>and</strong> the MGF <strong>for</strong> a r<strong>and</strong>om variable Xare closely related. The <strong>for</strong>mer is the expectation of t X whilst the latter is theexpectation of e tX :Φ X (t) =E [ t X] , M X (t) =E [ e tX] .The MGF can thus be obtained from the PGF by replacing t by e t , <strong>and</strong> viceversa. The MGF has more general applicability, however, since it can be usedwith both continuous <strong>and</strong> discrete distributions whilst the PGF is restricted tonon-negative integer distributions.As its name suggests, the MGF is particularly useful <strong>for</strong> obtaining the momentsof a distribution, as is easily seen by noting thatE [ e tX] = E[1+tX + t2 X 2 ]+ ···2!=1+E[X]t + E [ X 2] t 22! + ··· .Assuming that the MGF exists <strong>for</strong> all t around the point t = 0, we can deducethat the moments of a distribution are given in terms of its MGF by∣E[X n ]= dn M X (t) ∣∣∣t=0dt n . (30.86)Similarly, by substitution in (30.51), the variance of the distribution is given byV [X] =M ′′ X(0) − [ M ′ X(0) ] 2, (30.87)where the prime denotes differentiation with respect to t.◮The MGF <strong>for</strong> the Gaussian distribution (see the end of subsection 30.9.1) is given byM X (t) =exp ( µt + 1 2 σ2 t 2) .Find the expectation <strong>and</strong> variance of this distribution.Using (30.86),M X(t) ′ = ( µ + σ 2 t ) exp ( µt + 1 2 σ2 t 2) ⇒ E[X] =M X(0) ′ = µ,M X(t) ′′ = [ σ 2 +(µ + σ 2 t) 2] exp ( µt + 1 2 σ2 t 2) ⇒ M X(0) ′′ = σ 2 + µ 2 .Thus, using (30.87),V [X] =σ 2 + µ 2 − µ 2 = σ 2 .That the mean is found to be µ <strong>and</strong> the variance σ 2 justifies the use of these symbols inthe Gaussian distribution. ◭The moment generating function has several useful properties that follow fromits definition <strong>and</strong> can be employed in simplifying calculations.1163

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!