MSc. Econ: MATHEMATICAL STATISTICS, 1996
The Moment Generating Function of the Binomial Distribution
Consider the binomial function
(1) b(x; n, p) = n!
x!(n − x)!pxqn−x with q = 1 − p.
Then the moment generating function is given by
(2)
Mx(t) = Xn
x=0
ext n!
x!(n − x)!pxqn−x
= Xn
x=0
(pet
)
x n!
x!(n − x)! qn−x
= (q + pet
)
n,
where the final equality is understood by recognising that it represents the
expansion of binomial. If we differentiate the moment generating function with
respect to t using the function-of-a-function rule, then we get
(3)
dMx(t)
dt = n(q + pet
)
n−1pet
= npet
(q + pet
)
n−1.
Evaluating this at t = 0 gives
(4) E(x) = np(q + p)
n−1 = np.
Notice that this result is already familiar and that we have obtained it previously
by somewhat simpler means.
To find the second moment, we use the product rule
(5) duv
dx = u
dv
dx + v
du
dx
to get
(6)
d2Mx(t)
dt2 = npet
©
(n − 1)(q + pet
)
n−2pet
ª
+ (q + pet
)
n−1©
npet
ª
= npet
(q + pet
)
n−2©
(n − 1)pet + (q + pet
)
ª
= npet
(q + pet
)
n−2©
q + npet
ª
.
1
MSc. Econ: MATHEMATICAL STATISTICS: BRIEF NOTES, 1996
Evaluating this at t = 0 gives
(7) E(x2) = np(q + p)
n−2(q + np)
= np(q + np).
From this, we see that
(8)
V (x) = E(x2) − ©
E(x)
ª2
= np(q + np) − n2p2
= npq.
Theorems Concerning Moment Generating Functions
In finding the variance of the binomial distribution, we have pursed a
method which is more laborious than it need by. The following theorem shows
how to generate the moments about an arbitrary datum which we may take to
be the mean of the distribution.
(9) The function which generates moments about the mean of a random
variable is given by Mx−µ(t) = exp{−µt}Mx(t) where Mx(t)
is the function which generates moments about the origin.
This result is understood by considering the following identity:
(10) Mx−µ(t) = E
©
exp{(x − µ)t}
ª = e−µtE(ext) = exp{−µt}Mx(t).
For an example, consider once more the binomial function. The moment
generating function about the mean is then
(11)
Mx−µ(t) = e−npt(q + pet
)
n
= (qe−pt + pet
e−pt)
n
= (qe−pt + peqt)
n.
Differentiating this once gives
(12) dMx−µ(t)
dt = n(qe−pt + peqt)
n−1(−pqe−pt + qpeqt).
At t = 0, this has the value of zero, as it should. Differentiating a second time
according to the product rule gives
(13) d2Mx−µ(t)
dt2 = u(p2qe−pt + q2peqt) + v
du
dt ,
2
MSc. Econ: MATHEMATICAL STATISTICS, 1996
where
(14)
u(t) = n(qe−pt + peqt) and
v(t)=(−pqe−pt + qpeqt).
At t = 0 these become u(0) = n and v(0) = 0. It follows that
(15) V (x) = n(p2q + q2p) = npq(p + q) = npq,
as we know from a previous derivation.
Another important theorem concerns the moment generating function of
a sum of independent random variables:
(16) If x ∼ f(x) and y ∼ f(y) be two independently distributed
random variables with moment generating functions Mx(t) and
My(t), then their sum z = x+y has the moment generating function
Mz(t) = Mx(t)My(t).
This result is a consequence of the fact that the independence of x and y
implies that their joint probability density function is the product of their
individual marginal probability density functions: f(x, y) = f(x)f(y). From
this, it follows that
(17)
Mx+y(t) = Z
x
Z
y
e(x+y)t
f(x, y)dydx
=
Z
x
extf(x)dx Z
y
eytf(y)dy
= Mx(t)My(t).
ปริญญาโทจากชโรด: คณิตศาสตร์สถิติ 1996ในขณะที่สร้างฟังก์ชันของการแจกแจงทวินามพิจารณาฟังก์ชันทวินาม(1) b(x; n, p) = nx (n − x) ! pxqn−x กับ q = p − 1แล้ว เวลาสร้างฟังก์ชันถูกกำหนดโดย(2)Mx(t) = Xnx = 0ต่อ nx (n − x) ! pxqn−x= Xnx = 0(สัตว์เลี้ยง)x nx (n − x) qn−x= (q + สัตว์เลี้ยง)nซึ่งเป็นที่เข้าใจ โดยตระหนักถึงว่า มันแสดงถึงความเสมอภาคสุดท้ายการขยายตัวของทวินาม ถ้าเราแยกความแตกต่างในขณะที่สร้างฟังก์ชันด้วยเคารพโดยใช้กฎฟังก์ชันของ-a-ฟังก์ชัน t แล้วเรารับ(3)dMx(t)dt = n (q + สัตว์เลี้ยง)n−1pet= npet(q + สัตว์เลี้ยง)n−1ประเมินนี้ที่ t = 0 ให้(4) E(x) = np (q + p)n−1 = npสังเกตผลลัพธ์นี้อยู่แล้วคุ้นเคย และการที่ เราได้รับมันก่อนหน้านี้โดยวิธีค่อนข้างง่ายกว่าในการค้นหาขณะที่สอง เราใช้กฎผลิตภัณฑ์(5) duvdx = udvdx + vดูdxจะได้รับ(6)d2Mx(t)dt2 = npet©(n − 1) (q + สัตว์เลี้ยง)n−2petª(q + สัตว์เลี้ยง)n−1 ©npetª= npet(q + สัตว์เลี้ยง)n−2 ©(n − 1) สัตว์เลี้ยง (q + สัตว์เลี้ยง)ª= npet(q + สัตว์เลี้ยง)n−2 ©q + npetª.1ปริญญาโทจากชโรด: สถิติคณิตศาสตร์: บันทึกย่อ 1996ประเมินนี้ที่ t = 0 ให้(7) E(x2) = np (q + p)n−2(q + np)= np (q + np)จากนี้ เราเห็นว่า(8)V (x) = E(x2) − ©E(x)ª2= np (q + np) − n2p2= npqทฤษฎีที่เกี่ยวข้องกับช่วงเวลาที่สร้างฟังก์ชันในการหาผลต่างของการแจกแจงทวินาม เรามี pursedวิธีซึ่งลำบากมากกว่าจะตามได้ ทฤษฎีบทต่อไปนี้แสดงhow to generate the moments about an arbitrary datum which we may take tobe the mean of the distribution.(9) The function which generates moments about the mean of a randomvariable is given by Mx−µ(t) = exp{−µt}Mx(t) where Mx(t)is the function which generates moments about the origin.This result is understood by considering the following identity:(10) Mx−µ(t) = E©exp{(x − µ)t}ª = e−µtE(ext) = exp{−µt}Mx(t).For an example, consider once more the binomial function. The momentgenerating function about the mean is then(11)Mx−µ(t) = e−npt(q + pet)n= (qe−pt + pete−pt)n= (qe−pt + peqt)n.Differentiating this once gives(12) dMx−µ(t)dt = n(qe−pt + peqt)n−1(−pqe−pt + qpeqt).At t = 0, this has the value of zero, as it should. Differentiating a second timeaccording to the product rule gives(13) d2Mx−µ(t)dt2 = u(p2qe−pt + q2peqt) + vdudt ,2MSc. Econ: MATHEMATICAL STATISTICS, 1996where(14)u(t) = n(qe−pt + peqt) andv(t)=(−pqe−pt + qpeqt).At t = 0 these become u(0) = n and v(0) = 0. It follows that(15) V (x) = n(p2q + q2p) = npq(p + q) = npq,as we know from a previous derivation.Another important theorem concerns the moment generating function ofa sum of independent random variables:(16) If x ∼ f(x) and y ∼ f(y) be two independently distributedrandom variables with moment generating functions Mx(t) andMy(t), then their sum z = x+y has the moment generating functionMz(t) = Mx(t)My(t).This result is a consequence of the fact that the independence of x and yimplies that their joint probability density function is the product of theirindividual marginal probability density functions: f(x, y) = f(x)f(y). Fromthis, it follows that(17)Mx+y(t) = ZxZye(x+y)tf(x, y)dydx=Zxextf(x)dx Zyeytf(y)dy= Mx(t)My(t).
การแปล กรุณารอสักครู่..
