how to calculate the variance estimate of each row in r-CodePudding

I need to write a small program where I need to ask the user and enter the number of subintervals (2 or more). Divide the elements of the TT vector into a given number of connected, approximately equal subintervals. Calculate variance estimates on each subinterval. Display the results. Using the Bartlett criterion to test the hypothesis of equality of variances on all subintervals – with the alternative "not equal". The task is simple, but I do not know how to find out the variance estimate, at each interval and without a cycle. My code:

> TT
[1] 20.2 18.6 15.0 12.0 11.7 10.9  9.0 11.9 13.3  8.8  8.6  6.1  6.6  6.5 11.4
[16] 12.9  5.4  2.5  4.3  3.0

> n <- as.numeric(readline(prompt = "Enter court intervals: "))
Enter court intervals: 3
> n
[1] 3
> ints = split(TT, cut(seq_along(TT),n))
> ints
$`(0.981,7.33]`
[1] 20.2 18.6 15.0 12.0 11.7 10.9  9.0

$`(7.33,13.7]`
[1] 11.9 13.3  8.8  8.6  6.1  6.6

$`(13.7,20]`
[1]  6.5 11.4 12.9  5.4  2.5  4.3  3.0

And then I don't know how to calculate the variance estimate for each interval without a cycle

CodePudding user response：

If by "cycle", you mean a loop, use lapply:

variance <- lapply(ints, var) # Or sapply(ints, var)