2.

6 The Poisson Distribution 習題第 1 頁共 3 頁

2.6-2 Let X have a Poisson distribution with a variance of 5. Find P ( X  3) .

53  e 5 125e 5
解： P( X  3)   .
3! 6

2.6-3 Customers arrive at a travel agency at a mean rate of 11 per hour. Assuming that the number of
arrivals per hour has a Poisson distribution, give the probability that more than 10 customers arrive in a
given hour.
解： X ~ P(11) ，查卜瓦松分配表
P( X  10)  1  P( X  10)  1  0.460  0.540

2.6-4 If X has a Poisson distribution such that P( X  0)  P( X  1) . Find the mean of X .

𝜆0 𝑒 −𝜆 𝜆1 𝑒 −𝜆
解： =
0! 1!

𝑒 −𝜆 (𝜆 − 1) = 0
𝜆=1
𝐸 (𝑋 ) = 𝜆 = 1

2.6-5 Flaws in a certain type of drapery(布匹) material appear on the average of one in 150 square feet. If we
assume a Poisson distribution, find the probability of at most one flaw appearing in a 225 square feet.
解： X ~ P(1.5)
1
1.5 x  e 1.5
P( X  1)    2.5e 1.5  0.558
x 0 x!

2.6-6 A certain type of aluminum screen that is 4 feet wide has, on the average, one flaw in an 85-foot roll.
Find the probability that a 35-foot roll has no flaws.
解： X ~ P(35 / 85)
0
 35   85
35

  e 7
P( X  0)   
85 
 e 17
0!

2.6-7 With probability 0.001 , a prize of $499 is won in the Michigan Daily Lottery when a $1 straight bet is
placed. Let Y equal the number of $499 prizes won by a gambler after placing n straight bets. Note that
Y is b(n, 0.001) . After placing n  2000 $1 bets, a gambler is behind or even if {Y  4} . Use the
Poisson distribution to approximate P(Y  4) when n  2000 .
解： Y ~ b(2000, 0.001)
𝜆 = 2000(0.001) = 2，則𝑋近似𝑃(2)，查卜瓦松分配表
𝑃(𝑌 ≤ 4) = 0.947
 2.6 The Poisson Distribution 習題第 2 頁共 3 頁
2.6-9 A store selling newspapers only orders n  4 of a certain newspaper because the manager does not
get many calls for that publication. If the number of requests per day follows a Poisson distribution with
mean 3.
(a) What is the expected value of the number sold?
(b) What is the minimum number that the manager should order so that the chance of having more
requests than available newspapers is less than 0.05 ?
解：(a) 𝑋~𝑃(3)，查卜瓦松分配表
(0)𝑃 (𝑋 = 0) + (1)𝑃(𝑋 = 1) + (2)𝑃 (𝑋 = 2) + (3)𝑃 (𝑋 = 3) + (4)𝑃 (𝑋 ≥ 4)
= [𝑃(𝑋 ≤ 1) − 𝑃(𝑋 ≤ 0)] + 2[𝑃(𝑋 ≤ 2) − 𝑃(𝑋 ≤ 1)] + 3[𝑃(𝑋 ≤ 3) − 𝑃(𝑋 ≤ 2)] + 4[1 − 𝑃(𝑋 ≤ 3)]
= 4 − 𝑃(𝑋 ≤ 0) − 𝑃 (𝑋 ≤ 1) − 𝑃 (𝑋 ≤ 2) − 𝑃 (𝑋 ≤ 3)
= 4 − 0.050 − 0.199 − 0.423 − 0.647 = 2.681

(b) 訂 5 份報紙時: P( X  5)  1  P( X  5)  1  0.916  0.084  0.05

訂 6 份報紙時: P( X  6)  1  P( X  6)  1  0.966  0.034  0.05
因此，訂 6 份報紙即可。

2.6-10 The mean of a Poisson random variable X is   9 . Compute P(   2  X    2 ) .

解： X ~ P(9)
𝜎2 = 9
𝜎 = 3 (∵ 𝜎 > 0)
P(   2  X    2 )  P[9  2(3)  X  9  2(3)]  P(3  X  15)  P(4  X  14)
 P( X  14)  P( X  3)  0.959  0.021  0.938

2.6-11 An airline always overbooks(超量預訂) if possible. A particular plane has 95 seats on a flight in
which a ticket sells for $300. The airline sells 100 such tickets for this flight.
(a) If the probability of an individual not showing up (到場登機) is 0.05, assuming independence, what
is the probability that the airline can accommodate(可搭載) all the passengers who do show up? (請用
卜瓦松分配求近似機率值)
(b) If the airline must return the $300 price plus a penalty(罰金) of $400 to each passenger that cannot get
on the flight, what is the expected payout (penalty plus ticket refund) that the airline will pay? (請用卜
瓦松分配求近似值)
解：令 X 代表 100 位預訂機位者，未到場登機的人數，則 X ~ b(100, 0.05) 。
𝜆 = 100(0.05) = 5，則𝑋近似𝑃 (5)。
(a) 𝑃 (𝑋 ≥ 5) = 1 − 𝑃 (𝑋 ≤ 4) = 1 − 0.440 = 0.560
(b) 700[5𝑃 (𝑋 = 0) + 4𝑃 (𝑋 = 1) + 3𝑃 (𝑋 = 2) + 2𝑃 (𝑋 = 3) + 1𝑃(𝑋 = 4)]
= 700{5𝑃(𝑋 ≤ 0) + 4[𝑃(𝑋 ≤ 1) − 𝑃(𝑋 ≤ 0)] + 3[𝑃(𝑋 ≤ 2) − 𝑃(𝑋 ≤ 1)] + 2[𝑃(𝑋 ≤ 3) − 𝑃(𝑋 ≤ 2)] + 𝑃(𝑋 ≤ 4) − 𝑃(𝑋 ≤ 3)}

= 700[𝑃 (𝑋 ≤ 0) + 𝑃(𝑋 ≤ 1) + 𝑃(𝑋 ≤ 2) + 𝑃 (𝑋 ≤ 3) + 𝑃(𝑋 ≤ 4)]

= 700(0.007 + 0.040 + 0.125 + 0.265 + 0.440) = 613.9
 2.6 The Poisson Distribution 習題第 3 頁共 3 頁
2.6-12 A baseball team loses $88000 for each consecutive day it rains. Say 𝑋, the number of consecutive
days it rains at the beginning of the season, has a Poisson distribution with mean 0.32. What is the
expected loss before the opening game?
解：88000𝐸 (𝑋) = 88000(0.32) = 28160

2.6-13 Assume that a policyholder is four times more likely to file exactly two claims (申請 2 件理賠) as to
file exactly three claims. Assume also that the number 𝑋 of claims of this policyholder is Poisson.
Determine the expectation 𝐸 (𝑋 2 ).
𝜆2 𝑒 −𝜆 𝜆3 𝑒 −𝜆
解： =4∙
2! 3!

4𝜆
𝜆2 𝑒 −𝜆 ( 3 − 1) = 0

3
𝜆 = 0 (不合，因𝜆 > 0) 或 𝜆 = 4

Var(𝑋) = 𝐸 (𝑋 2 ) − 𝜇 2
3 3 2
= 𝐸 (𝑋 2 ) − (4)
4

3 3 2 21
𝐸 (𝑋 2 ) = + ( ) =
4 4 16

參考資料:卜瓦松分配，𝑿~𝑷(𝝀)，𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙)
𝝀 = 𝑬(𝑿)
𝒙 0.5 1 1.5 2 3 4 5 9 11
0 0.607 0.368 0.223 0.135 0.050 0.018 0.007 0.000 0.000
1 0.910 0.736 0.558 0.406 0.199 0.092 0.040 0.001 0.000
2 0.986 0.920 0.809 0.677 0.423 0.238 0.125 0.006 0.001
3 0.998 0.981 0.934 0.857 0.647 0.433 0.265 0.021 0.005
4 1.000 0.996 0.981 0.947 0.815 0.629 0.440 0.055 0.015
5 1.000 0.999 0.996 0.983 0.916 0.785 0.616 0.116 0.038
6 1.000 1.000 0.999 0.995 0.966 0.889 0.762 0.207 0.079
7 1.000 1.000 1.000 0.999 0.988 0.949 0.867 0.324 0.143
8 1.000 1.000 1.000 1.000 0.996 0.979 0.932 0.456 0.232
9 1.000 1.000 1.000 1.000 0.999 0.992 0.968 0.587 0.341
10 1.000 1.000 1.000 1.000 1.000 0.997 0.986 0.706 0.460
11 1.000 1.000 1.000 1.000 1.000 0.999 0.995 0.803 0.579
12 1.000 1.000 1.000 1.000 1.000 1.000 0.998 0.876 0.689
13 1.000 1.000 1.000 1.000 1.000 1.000 0.999 0.926 0.781
14 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.959 0.854