来自《耶鲁大学开放课程：金融市场 第2集 第2集》

• 
  
  i = 1 to n of (x - x bar)2/n.
  x减去x均值的平方 除以n的累加
• 
  
  That is the sample variance.
  这就是样本方差
• 
  
  Some people will divide by n–1.
  另一种用法里 分母是n-1
• 
  
  I suppose I would accept either answer.
  我觉得两个都可以接受
• 
  
  I'm just keeping it simple here.
  这里想说的简单一点
• 
  
  They divide by n-1 to make it an unbiased estimator
  当除以n-1表示的是对总体的
• 
  
  of the population variance,
  无偏估计
• 
  
  but I'm just going to show it in a simple way here.
  我在这里只是说的简单一点
• 
  
  So you see what it is--it's a measure of
  你会看到 这个指标
• 
  
  how much x deviates from the mean,
  衡量的是x与平均值的偏离
• 
  
  but it's squared.
  而且这里有个平方
• 
  
  It weights big deviations a lot
  使偏离的权重更大