英文:
Variance over a 3D array in R
问题
在处理一个三维数组：
ar3d <- array(floor(100 * runif(100 * 20 * 30)), dim = c(100, 20, 30))
我感到困惑的是：
length(apply(ar3d, MARGIN = 1, FUN = "sd"))
# [1] 100
而
dim(apply(ar3d, MARGIN = 1, FUN = "var"))
# [1] 900 100
如何尽快地计算20 * 30个值之间的方差？
非常感谢！
英文:
Working on a 3d array:
ar3d <- array(floor(100 * runif(100 * 20 * 30)), dim = c(100, 20, 30))
I am confused why
length(apply(ar3d, MARGIN = 1, FUN = "sd"))
# [1] 100
While
dim(apply(ar3d, MARGIN = 1, FUN = "var"))
# [1] 900 100
How can I calculate the variance amongst the 20 * 30 values as fast as possible?
Many thanks
答案1
得分: 2
对于 var，如果你想获得一个标量而不是协方差矩阵，你应该使用 var(c(x))，而不是 var(x)。
当你键入 ?var 时，你会看到：
var、cov 和 cor 计算 x 的方差以及 x 和 y 的协方差或相关性，如果它们是向量的话。如果 x 和 y 是矩阵，那么计算 x 的列与 y 的列之间的协方差（或相关性）。
你可以像这样尝试：
> ar3d <- array(floor(100 * runif(100 * 20 * 30)), dim = c(100, 20, 30))
> length(apply(ar3d, 1, \(x) var(c(x))))
[1] 100
> length(apply(ar3d, 1, var))
[1] 90000
> length(apply(ar3d, 1, sd))
[1] 100
英文:
For var, you should use var(c(x)) instead of var(x) if you want to obtain a scalar rather a covariance matrix.
When you type ?var, you see
> var, cov and cor compute the variance of x and the covariance or
> correlation of x and y if these are vectors. If x and y are matrices
> then the covariances (or correlations) between the columns of x and
> the columns of y are computed.
You can try it out like
> ar3d <- array(floor(100 * runif(100 * 20 * 30)), dim = c(100, 20, 30))
> length(apply(ar3d, 1, \(x) var(c(x))))
[1] 100
> length(apply(ar3d, 1, var))
[1] 90000
> length(apply(ar3d, 1, sd))
[1] 100
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论