是否在标准化中知道方差,而在学生化中却不知道并据此估算?谢谢。
是否在标准化中知道方差,而在学生化中却不知道并据此估算?谢谢。
Answers:
简短回顾。给定一个模型,其中,X是Ñ × p,β = (X ' X )- 1 X ' Ŷ和ÿ = X β = X (X ' X )- 1 X ' ÿ = ħ ÿ,其中ħ = X (X ' X是“帽子矩阵”。残差是 ë = Ŷ - Ŷ = ý - ħ Ý = (我- ħ )ý 总体方差 σ 2是未知的,并且可以估算中号小号Ë,均方误差。
半学习残差定义为 但是,由于残差的方差取决于两个σ2和X,它们的估计方差是 :V(È我)=中号小号ë(1-ħ我我) 其中ħ我我是我个对角元素帽子矩阵。
标准化残差,也称为内部学生残差,为:
但是,单个和M S E是非独立的,因此r i不能具有t分布。该过程然后删除我个观察,拟合回归线功能向剩余Ñ - 1个观察,并得到新的ÿ其可以通过被表示的ÿ我(我)。差: d 我 = ÿ 我- ÿ我(我) 被称为
See Kutner et al., Applied Linear Statistical Models, Chapter 10.
Edit: I must say that the answer by rpierce is perfect. I thought that the OP was about standardized and studentized residuals (and dividing by the population standard deviation to get standardized residuals looked odd to me, of course), but I was wrong. I hope that my answer can help someone even if OT.
In social sciences it is typically said that Studentizated scores uses Student's/Gosset's calculation for estimating the population variance/standard deviation from the sample variance/standard deviation (). In contrast, Standardized scores (a noun, a particular type of statistic, the Z score) are said to use the population standard deviation ?().
However, it appears there is some terminological differences across fields (please see the comments on this answer). Therefore, one ought to proceed with caution in making these distinctions. Moreover, studentized scores are rarely called such and one typically sees 'studentized' values in the context of regression. @Sergio provides details about those types of studentized deleted residuals in his answer.
我很迟才回答这个问题!但是找不到非常简单的语言的答案,所以请谦虚地尝试回答。
我们为什么要标准化?假设您有两个模型,一个模型通过研究统计数据所花费的时间来预测疯狂,而另一个模型则通过统计数据所花费的时间来预测对数(疯狂)。
很难理解残差都在不同的单位中。因此我们将它们标准化。(与Z分数类似的理论)
标准化残差:-将残差除以标准差的估计值。通常,如果绝对值> 3,则值得关注。
我们用它来研究模型中的异常值。
学生化残差:我们用它来研究模型的稳定性。
过程很简单。我们从模型中删除单个测试用例,并找出新的预测值。可以通过除以标准误差来标准化新值与原始观测值之间的差异。此值为学生剩余数
有关使用R发现静态的更多信息-http: //www.statisticshell.com/html/dsur.html
Wikipedia在https://en.wikipedia.org/wiki/Normalization_(statistics)上有很好的概述:
标准分数 :在已知填充参数时归一化错误。适用于正态分布的人群
学生的t统计量 :在总体参数未知(估计)时归一化残差。