Displaying 1 result from an estimated 1 matches for "ordertim".
Did you mean:
ordertime
2005 Mar 24
1
Histogram over times (without dates)
...I want to know at what times on a day a costumer buys anything (times are collected from logfiles).
The values span 24 hours (e.g. 00:00:00 to 23:59:59)
I converted the characters to chron objects (library(chron)) and then to numeric vectors:
ordertm.num <- as.numeric(chron(times = ecom$Ordertime))
Then I put the numeric values to hist(), printed the hist without axes=F and constructed my own axis.
The result is satisfactory. But as there are so many possibilities with times on R (namely using the POSIX classes) I want to be shure if there is no "standard" approach to...