用户工具

站点工具


中心极限定律

差别

这里会显示出您选择的修订版和当前版本之间的差别。

到此差别页面的链接

两侧同时换到之前的修订记录前一修订版
后一修订版
前一修订版
中心极限定律 [2024/03/15 04:53] 2104龚文滕中心极限定律 [2024/03/15 04:57] (当前版本) 2104龚文滕
行 1: 行 1:
      
-**标准误(standard of error)**指样本均值分布的标准差,是反映样本均值分布变异性的指标。+**标准误(standard of error)** 
 +  *指样本均值分布的标准差,是反映样本均值分布变异性的指标。
   *定义式为σ/√n,σ为总体的标准差,n为样本容量。   *定义式为σ/√n,σ为总体的标准差,n为样本容量。
      
行 6: 行 7:
   *The definition formula is σ/√n, σ is the standard deviation of the population, and n is the sample size.   *The definition formula is σ/√n, σ is the standard deviation of the population, and n is the sample size.
    
-   +**大数定律(law of large numbers)** 
-**大数定律(law of large numbers)**随样本容量的的增大,样本均值与总体均值之间的误差会减小。 +  *随样本容量的的增大,样本均值与总体均值之间的误差会减小。 
-   +  *Law of large numbers: As the sample size increases, the error between the sample mean and the population mean will decrease.  
-  *Law of large numbers: As the sample size increases, the error between the sample mean and the population mean will decrease. +
-  +
 **中心极限定律(central limit theorem)** **中心极限定律(central limit theorem)**
   *对于任何均值为μ,标准差为σ的总体, 样本容量为n的样本均值的分布,随着n趋近无穷大时,会趋近均值为μ,标准差为σ/√n的正态分布 。   *对于任何均值为μ,标准差为σ的总体, 样本容量为n的样本均值的分布,随着n趋近无穷大时,会趋近均值为μ,标准差为σ/√n的正态分布 。
行 16: 行 15:
   *中心极限定律综合了样本均值的三个主要特性:形状、均值和方差。   *中心极限定律综合了样本均值的三个主要特性:形状、均值和方差。
  
-  *For any population with a mean of and a standard deviation of u, and a sample size of n, the distribution of the mean will approach a normal distribution with a mean of and a standard deviation of u/√ n as n approaches infinity.+  *For any population with a mean of μ and a standard deviation of σ, and a sample size of n, the distribution of the mean will approach a normal distribution with a mean of μ and a standard deviation of σ/√n as n approaches infinity
 +  *Application conditions: Usually, when n ≥ 30, we believe that the distribution of sample mean satisfies the central limit theorem. 
 +  *The central limit theorem combines the three main characteristics of sample mean: shape, mean, and variance.
中心极限定律.1710478424.txt.gz · 最后更改: 2024/03/15 04:53 由 2104龚文滕