在看一篇文章 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2752127/
里面方法部分有一句话,不是很明了:
we first convert the read count of a window into a Z-score zi by subtracting
the mean of all windows and dividing by the standard deviation. The Z-score
is then converted to its upper-tail probability p_Upperi = P(Z>zi), and its
lower-tail probability p_Loweri = P(Z
这里说把z score转化成probability,搞不懂。我知道怎么把z score转化成
percentage,因为+/- z score到0之间的面积百分比在 normal distribution里面有表
格可以转换。但是它这里的upper and lower probability到底是个什么东西?
我猜想,lower probability 是负无穷到- z score的面积百分比(也就是1-
percentage/2,这里的percentage是+/- z score到0之间的面积百分比),而upper
probability 是+ z score到正无穷的面积百分比?
求指教,多谢!
因为文章的HTML页面里面某个公式有错误,跟PDF不同,我都不敢肯定其他地方是不是
也有错误。
人家定义不是都写出来了
upper-tail probability p_Upperi = P(Z>zi)
lower-tail probability p_Loweri = P(Z
subtracting
score
its
【在 A**H 的大作中提到】 : 在看一篇文章 : http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2752127/ : 里面方法部分有一句话,不是很明了: : we first convert the read count of a window into a Z-score zi by subtracting : the mean of all windows and dividing by the standard deviation. The Z-score : is then converted to its upper-tail probability p_Upperi = P(Z>zi), and its : lower-tail probability p_Loweri = P(Z: 这里说把z score转化成probability,搞不懂。我知道怎么把z score转化成 : percentage,因为+/- z score到0之间的面积百分比在 normal distribution里面有表 : 格可以转换。但是它这里的upper and lower probability到底是个什么东西?