Binomial distribution

Japanese: 二項分布 - にこうぶんぷ
Binomial distribution

Let p be the probability that event E occurs in a certain trial. If the trial is repeated independently n times and the number of times E occurs is denoted as X, then X is a random variable, and the probability that X=k, p k =P(X=k), is given by the following equation, where q=1-p.

p k = n C k p k q n k
(k=0,1,……,n)
This probability distribution is called the binomial distribution and is represented as B(n,p).

[Example 1] If you roll a dice four times, what is the probability of getting a six exactly two times?

It is.

The above nC k p k q nk is the formula n C 0 q n + n C 1 pq n-1 +…… which is obtained by expanding (q+p) n by the binomial theorem.
+ n C k p k q nk +……+ n C n p n
It is a term. If we create a line graph of the binomial distribution B(n,p) with k on the horizontal axis and p k on the vertical axis (k=0,1,2,……,n), we get the following. If k 0 =np+p-1 is an integer, then p k monotonically increases when k<k 0 , reaches its maximum value when k=k 0 ,k 0 +1, and monotonically decreases when k>k 0 +1. If np+p-1 is not an integer, then if the integer part of np+p-1 is k 0 , then p k monotonically increases when k<k 0 , reaches its maximum value when k=k 0 , and monotonically decreases when k>k 0. The following can be derived from the definition of binomial distribution mentioned at the beginning. The random variables X 1 , …, X n are independent and P(X i =1)=p, P(X i =0)=q=1-p
(i=1,……,n)
In this case, the random variable X 1 +X 2 +……+X n
The probability distribution of is the binomial distribution B(n,p). The mean of the binomial distribution B(n,p) is np, the variance is npq, and the characteristic function is (pe it +q) n .

[Example 2] When a dice is rolled 500 times, find the probability p that the number 1 will appear 80 or more times.

The desired probability p is

However, it is not easy to calculate this value directly. For problems of this type, we use a normal distribution table as follows. Approximation of the binomial distribution by normal distribution When the distribution of the random variable X is binomial distribution B(n,p), if n is large, the random variable

The distribution of is close to the standard normal distribution (Laplace's theorem). Using this, we can approximate the probability p in the previous example (Example 2). For two integers a and b, 0≦a<b≦n, p(a≦X≦b) is the area of ​​the blue part in the figure , which is the graph of the normal distribution N(np,np(1-p)) with the same mean and variance as X, the x-axis, and the two straight lines x=a-0.5, x=b+0.5
The area of ​​the area enclosed by the square root of the square root is approximately equal to the area of ​​the area enclosed by the square root of the square root. Therefore, if we assume that the distribution of Z is a standard normal distribution,

Then, the following approximation holds:

P(a≦X≦b)≒P(α≦Z≦β)
Using this formula and a normal distribution table to calculate p for Example 2, we get p = 0.67.

[Shigeru Furuya]

Example of binomial distribution (figure)
©Shogakukan ">

Example of binomial distribution (figure)


Source: Shogakukan Encyclopedia Nipponica About Encyclopedia Nipponica Information | Legend

Japanese:

ある試行において事象Eのおこる確率をpとする。この試行を独立にn回繰り返す場合にEのおこる回数をXと置くと、Xは確率変数であって、X=kとなる確率pk=P(X=k)は、q=1-pとして次式で与えられる。

  pk=nCkpkqn-k
     (k=0,1,……,n)
この確率分布を二項分布といい、B(n,p)と表す。

〔例1〕さいころを4回投げて、そのうちちょうど2回6の目の出る確率は

である。

 前記のnCkpkqn-kは、(q+p)nを二項定理によって展開した式
  nC0qn+nC1pqn-1+……
   +nCkpkqn-k+……+nCnpn
の項になっている。二項分布B(n,p)において、kを横軸に、pkを縦軸にとって(k=0,1,2,……,n)、折れ線グラフをつくると次のようになる。k0=np+p-1が整数であれば、pkはk<k0で単調増加、k=k0,k0+1で最大値をとり、k>k0+1で単調減少である。np+p-1が整数でなければ、np+p-1の整数部分をk0とするとpkはk<k0で単調増加で、k=k0で最大値をとり、k>k0で単調減少である。初めに述べた二項分布の定義から次のことが導かれる。確率変数X1、……、Xnは独立で
  P(Xi=1)=p, P(Xi=0)=q=1-p
  (i=1,……,n)
とする。このとき確率変数
  X1+X2+……+Xn
の確率分布は二項分布B(n,p)である。二項分布B(n,p)の平均値はnp、分散はnpqであり、特性関数は(peit+q)nである。

〔例2〕さいころを500回投げるとき、1の目が80回以上出る確率pを求めよ。

 求める確率pは

であるが、この値を直接計算することは容易でない。このような形の問題に対しては、次のように正規分布表を用いて計算を行う。二項分布の正規分布による近似確率変数Xの分布が二項分布B(n,p)であるとき、nが大きければ、確率変数

の分布は標準正規分布に近い(ラプラスの定理)。このことを利用して前の〔例2〕の確率pの近似値を求めることができる。0≦a<b≦nである二つの整数a、bに対して、p(a≦X≦b)はの青色部分の面積で、これはXと同じ平均および分散をもつ正規分布N(np,np(1-p))のグラフとx軸および2直線
  x=a-0.5, x=b+0.5
で囲まれた部分の面積とほぼ等しい。したがって、Zの分布を標準正規分布として

と置けば、次の近似式が成り立つ。

  P(a≦X≦b)≒P(α≦Z≦β)
この式により正規分布表を用いて〔例2〕のpを求めるとp=0.67が得られる。

[古屋 茂]

二項分布の例〔図〕
©Shogakukan">

二項分布の例〔図〕


出典 小学館 日本大百科全書(ニッポニカ)日本大百科全書(ニッポニカ)について 情報 | 凡例

<<:  Nikogori (boiled and condensed food) - Nikogori

>>:  Binomial theorem

Recommend

Myxedema - Myxedema

A disease caused by hypothyroidism. Most cases are...

Farming Rites - Hatasakugirei

Agricultural rites associated with field crops. Be...

Tourte

…The violin, which had established a firm positio...

Fractal - Furakutaru

In nature, irregular and complex shapes exist eve...

Iga Kokubunji Temple

...Former name of province. Ishu. Currently, it i...

yippie

...As a result, hippies gradually became a symbol...

Negroponte

…Population: 210,000 (1991). Also called Évvoia o...

Pipe tree

…It was introduced to Japan in the middle of the ...

Kanzanchiku - Kanzanchiku

A bamboo or bamboo species of the grass family (A...

Salt seasoning - Kanmiryo

…Also, because the spiciness is more of a physica...

Cynoglossus robustus (English spelling)

…They are found in the shallow seabeds of tempera...

Ainsworth, GC - Ainsworth

…By this time, mycology had become an independent...

Hamada Hirosuke - Hamada Hirosuke

A children's writer. Born on May 25, 1893 in ...

Kanada [town] - Kanada

An old town in Tagawa County, in the north-central...

Rokkaku-shi

Shugo of Omi Province and a feudal lord during th...