Finding the index values for a combination of vectors-CodePudding

In my reproducible data below, I wonder how to find the index values for elements from the one-on-one matched vectors rs & hat that satisfy the following condition:

rs < -1.96 OR rs > 1.96 elements whose corresponding hat elements are hat > max_hat elements

Is this possible to plot (using text()) the index values next to its plotted circle in the plot below?

x <- readRDS(url("https://github.com/rnorouzian/s/raw/main/cooks.rds"))
hat <- readRDS(url("https://github.com/rnorouzian/s/raw/main/hat.rds"))
rs <- readRDS(url("https://github.com/rnorouzian/s/raw/main/rstudent.rds"))$z


plot(hat, rs, cex = 0.2 3*sqrt(x), pch = 19)

max_hat <- boxplot.stats(hat, coef = 2)$stats[5]

CodePudding user response：

Both conditions define logical vectors so AND them and get which indices they correspond to. Then plot.
I have added color to make the text labels more obvious.

max_hat <- boxplot.stats(hat, coef = 2)$stats[5]

i <- abs(rs) > 1.96  # abs() because  /-1.96 are 
                     # symmetric values
j <- hat > max_hat
k <- which(i & j)

plot(hat, rs, cex = 0.2 3*sqrt(x), pch = 19)
text(hat[k], rs[k], labels = names(k), pos = 2, col = "red")