[R] Why 95% "quantile" empty in R or why 95% "quantile" = 1 with data values between 0 and 1?

[Uwe Ligges]

On 10.07.2012 02:55, Angelr wrote:
> I am calling quantiles as follows. I don't understand why sometimes the
> columns (data values) above 95% are returned as "NULL"!! When I drop the
> percentile down to 92%, I see colums appearing. Why would any quantile be
> empty? I see sometimes that 95% percentile is being chosen as "1" for my
> data between 0 and 1, where obviously there's no column value equal to 1.
> But why would 95% of a data be = 1 when there's clearly no value=1?


Depends on the definition of "quantile" which is not unique, and hence R 
knows t least 9 definitions in the quatile() function. See ?quantile and 
choose a definition that fits your expectation.

Best,
Uwe Ligges



> Thanks.
>
> makeQuantiles <- function(x) {
>    t(apply(x, 1, function(row) { quantile(row, probs = c(0.50,0.75, 0.90,
> 0.91,0.92,0.93,0.94,0.95,0.96,0.97,0.98)
> ) }))
> }
> colsGTQuantile <- function(x, qs, q) {
>    sapply(rownames(x), function(rn) {
>      cat(sprintf("*** row name = %s, q = %s, val = %f\n", rn, q, qs[rn, q]))
>      names(x[rn, x[rn,] > qs[rn, q]])
>    }, simplify = TRUE)
> }
>
>
> df4 <- colsGTQuantile(km$points, qs, "95%")
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/Why-95-quantile-empty-in-R-or-why-95-quantile-1-with-data-values-between-0-and-1-tp4635934.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>