二项式分布是否的概率分布。伯努利试验的结果,即如果伯努利试验进行了n次,则其成功的概率由二项式分布给出。请记住,每个试验都独立于其他试验,只有两个可能的结果满足伯努利试验的相同条件。
考虑抛硬币n次的情况,正好获得x否的概率。可以使用二项式分布来计算头/尾的数量。如果在相同情况下仅一次抛硬币与伯努利分布相同就抛硬币。
如果它的概率分布函数由下式给出,则取值1,2,… n的随机变量X遵循二项分布。
P(X = r)= n C r p r q nr
在哪里,
r = 0, 1,2……, n, where p, q>0 such that p+q=1
p = probability of success of an event
q = probability of failure of an event
二项式分布的均值或期望值
二项式分布的平均值与其他任何事物的平均值相同,等于提交的乘积为no。成功的概率和每次成功的概率。
Mean = ∑r r. P(r)
= ∑r r nCr pr qn-r
= ∑r r n/r n-1Cr-1 p.pr-1 qn-r [as nCr= n/r n-1Cr-1]
= np ∑r n-1Cr-1 pr-1 q(n-1)-(r-1)
= np(q+p)n-1 [by binomial theorem i.e. (a+b)n = ∑k=0 nCk an bn-k ]
=np [as p+q=1]
Therefore, Mean=np
二项式分布的方差
我们知道,方差是对数据从数据集中的平均值的分布程度的度量。同样,二项式分布的方差是对每个no概率分布的度量。从均值概率得出的成功率,均值是从均值的平方差的平均值。
Variance = (∑r r2. P(r)) – Mean2
= ∑r [r(r-1)+r] nCr pr qn-r – (np)2
= ∑r r(r-1) nCr pr qn-r + ∑r r nCr pr qn-r – (np)2
= ∑r r(r-1) n/r (n-1)/(r-1) n-2Cr-2 p2 pr-2 qn-r +np – (np)2
= n(n-1)p2 {∑r n-2Cr-2 pr-2 qn-r } +np – (np)2
= n(n-1) p2 (q+p)n-2 + np – n2p2 [by binomial theorem i.e. (a+b)n = ∑k=0 nCk an bn-k ]
= n2p2 -np2 +np-n2p2 [as p+q=1]
= np-np2
= np(1-p)
= npq
Therefore, Variance=npq
二项式分布的标准偏差
标准偏差也是找出No分布程度的标准度量。从平均值。
Standard Deviation = (Variance)1/2
= (npq)1/2
例子1.一枚硬币被扔了五次。正好达到3倍的机率是多少?还可以找到均值,方差和标准差。
解决方案:
n = 5 (no. of trials)
p = probability of getting head at each trial
=1/2
q = 1-1/2 = 1/2
r = 3 ( no. of successes i.e. getting a head)
P(X=r) = nCr pr qn-r
= 5C3 (1/2)3 (1/2)5-3
= 5!/(3!*2!) 1/8 * 1/4
= 10 *(1/8)*(1/4)
= 5/16
Mean = np
= 5 * 1/2 = 5/2
Variance = npq = 5 * 1/2 * 1/2
= 5/4
Standard deviation = (5/4)1/
= 51/2/2
例子2.一个骰子被扔三次。得到偶数的概率是多少。二项式分布的均值,方差和标准差是多少?
解决方案:
Here, n = 3(no. of trials)
p = probability of getting an even number during each trial
p = 3/6=1/2 [ 2,4,6 are even no. in dice]
q = 1-1/2 =1/2
r= 1( no. of successes i.e. getting a even no. )
P(X=r)= nCr * pr * qn-r
=3C1 (1/2)1 (1/2)3-1
= 3!/(1!*2!) 1/2 * 1/4
= 3 * (1/2) * (1/4)
= 3/8
Mean = np
= 3 * 1/2 = 3/2
Variance = npq = 3 * 1/2 * 1/2
= 3/4
Standard deviation= (3/4)1/2
=31/2/2
示例3.如果有缺陷螺栓的概率为0.1,则求出总共500个螺栓中有缺陷螺栓的分布的平均值,方差和标准偏差。
解决方案:
Considering as a case of binomial distribution ,
n = 500( no. of trials which we can are no. of bolts here)
p = probability of one defective bolt during each trial
p = 0.1
Q=1-0.1 =0.9
Mean = np
= 500 * 0.1 =50
Variance = npq
= 500 * 0.1 * 0.9 = 45
Standard Deviation = (variance)1/2
= (45)1/2 = 6.71
示例4.从一副52张纸牌中连续抽出两张纸牌进行替换。求出A数的概率分布。还可以找到均值,方差和标准差。
解决方案:
n = 2(no. of trials)
p = probability of getting an ace in each trial
= 4/52 =1/13
q = 1-1/13 =12/13
r = no. of successes i.e no. of aces (0,1,2)
P(X=r) = 2Cr (1/13)r (12/13)2-r
For r = 0
P(0) = 2C0 (1/13) (12/13) 2-0
= 144/169
For r=1
P(1) = 2C1 (1/13)1 (12/13)2-1
= 24/169
For r=2
P(2) = 2C2 (1/13) 2 (12/13)2-2
=1/169
Therefore, probability distribution can be given as :
X | 0 | 1 | 2 |
P(X) | 144/169 | 24/169 | 1/169 |
Mean =np
= 2 * 1/13 =2/13
Variance = npq
= 2 * (1/13) * (12/13)
= 24/16
Standard Deviation= (variance)1/2
= (24/169)1/2 = 0.376