Multinomial Distribution（多项式分布）

DEDSEC_Roger

已于 2022-02-01 16:39:32 修改

阅读量1.7k

点赞数

分类专栏：深度学习文章标签：深度学习

于 2021-12-03 16:17:09 首次发布

本文链接：https://blog.csdn.net/m0_46324847/article/details/121700083

版权

深度学习专栏收录该内容

6 篇文章

订阅专栏

多项式分布是二项式分布的推广，允许每次试验有多于两种可能的结果。在多项式分布中，每个独立试验有k种互斥的可能结果，每种结果发生的概率分别为p_1, p_2, ..., p_k，且这些概率之和为1。随机变量X_i表示第i种结果发生的次数，它们满足0到n的约束条件，即所有结果的次数总和等于n。每个X_i单独看都有一个二项式分布，其期望和方差分别为E(X_i)=npi和Var(X_i)=npi(1-p_i)。举例来说，在美国的血型分布中，抽取10人的样本中，可以使用多项式分布来计算特定血型组合的概率。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

The multinomial distribution is a generalization of the binomial distribution.

In the binomial distribution there are only two possible outcomes on any one individual trial, and we labeled those success and failure. In the multinomial distribution, the number of possible outcomes on any one given trial is allowed to be greater than 2.

Suppose

There are n independent trials.
Each trial results in 1 of k mutually exclusive outcomes.
On any single trial these k outcomes comes occur with probabilities p_1, p_2 … p_k.

$\sum_{i=1}^{k}{p_{i}}=1$

Then

let the random variable X_i represent the number of occurrences of outcome of i.（让随机变量X_i表示i的结果的出现次数。）And i is going to take on the values 1 through k, representing those k possible outcomes on any one individual trial.（i的值为1到k，代表每次独立实验会出现的k种不同的结果。）
So we’re going to have k random variables, representing a count for each of those possible outcomes.
$P(X_{1}=x_{1},\cdot \cdot \cdot ,X_{k}=x_{k})=\frac{n!}{x_{1}!\cdot \cdot \cdot x_{k}!}p_{1}^{x_{1}}\cdot \cdot \cdot p_{k}^{x_{k}}$
p_i represent the probability of outcome i on any one individual trial
x_i represent the occurrences of outcome i within n individual trials
The random variable X_i can take on the possible values 0, 1, 2, …, n, and we know that n things must happen in total. So:
$\sum_{i=1}^{k}{x_{i}}=n$
And if we think about this a little bit, any one of these random variables, when viewed individually, it will have a binomial distribution. And if you remember our mean and variance for the binomial distribution:
$E(X_{i})=np_{i}\\ Var(X_{i})=np_{i}(1-p_{i})$

Example

This is approximately the distribution of blood types in the United States.

Type	O	A	B	AB
Probability	0.44	0.42	0.10	0.04

In a random sample of 10 Americans what is the probability 6 have blood type O, 2 have type A, 1 has type B, and 1 has type AB?
When any one individual person is sampled, they’re going to have one of these four blood types, according to these probabilities. And we’re going to be able to answer this question using the multinomial distribution.
$P(X_{1}=6, X_{2}=2 , X_{3}=1 , X_{4}=1)=\frac{10!}{6! 2! 1! 1!}0.44^{6}0.42^{2}0.10^{1}0.04^{1}$