duplicate row.names (index) error in zoo

Question

I have encountered a duplicate index warning/error in zoo. I spent some time to troubleshoot, and it came down to the following data/snippet:

Lines = "Index,dbt,rh,dpt,wind.mph,gust.mph,wind.dir,voltage,klux,fc
2008-03-09 01:00:00,37.96,77.25,31.60,4.16,5.39,247,0.00,28.283,2643.28
2008-03-09 02:00:00,37.20,77.25,30.90,0.00,0.43,244,0.00,28.283,2643.28"

x <- read.table(header=T,text=Lines,sep=",")

library(zoo)

z = zoo(x[,2:10],as.POSIXct(x$Index))

If I run the above, I got the following warning:

Warning message:
In zoo(x[, 2:10], as.POSIXct(x$Index)) :
  some methods for “zoo” objects do not work if the index entries in ‘order.by’ are not unique

I was wondering what is wrong with the data/index. They are in fact different.

However, the following command:

library(xts)
make.index.unique(index(z))

gave a puzzling result:

[1] "2008-03-09 01:00:00 MST" "2008-03-09 01:00:00 MST"

Did the above command just make the indeces duplicates? Can anybody enlighten me on how this works?

Answer 1

The issue is not so much with zoo as it is with the arguments to as.POSIXct

Have a look at as.POSIXct(x$Index) and you will see that it is not giving the results you require.

Giving as.POSIXct a time zone will help.

Compare:

as.POSIXct(x$Index)
#  [1] "2008-03-09 01:00:00 EST" "2008-03-09 01:00:00 EST"

as.POSIXct(x$Index, tz = "GMT")
#   [1] "2008-03-09 01:00:00 GMT" "2008-03-09 02:00:00 GMT"