[R] x12 doesn't work any longer
Rolf Turner
r.turner at auckland.ac.nz
Mon Oct 14 22:04:37 CEST 2013
On 10/15/13 02:52, Victor wrote:
> Context: Mac OS X 10.7 - R 3.0.2 GUI 1.62 Snow Leopard build (6558)
>
> I had been using x12 for some time with no problem at all. every month I update the time series ricpesata (see below) and run the simple procedure below. Now adding the data of the last month, Sep.2013,
> x12 is complaining:
>
>> library(x12)
>> x<-x12(round(ricpesata,2), x12path="/Applications/x12arima/x12a", automdl=TRUE, decimals=2, transform="log", outlier="all", forecast_years=2)
> Error in .local(object, x12Parameter, x12BaseInfo, ...) :
> unused arguments (x12path = "/Applications/x12arima/x12a", automdl = TRUE, decimals = 2, transform = "log", outlier = "all", forecast_years = 2)
>
> where
>> str(round(ricpesata,2))
> Time-Series [1:285] from 1990 to 2014: 20924 20559 19952 19366 19145 ...
>>
> is the following timeseries:
>
> Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
> 1990 20924.30 20559.41 19952.28 19365.96 19145.28 19536.69 19791.74 15796.91 19690.72 19779.16 20444.38 20529.32
> 1991 21155.47 21613.96 20168.89 19584.45 19527.75 19765.04 20314.15 16116.04 20232.00 20391.31 21200.69 21093.83
> 1992 21570.88 21877.04 21129.40 .........
>
> What does that error mean
Presumably it means what it says.
> and what should I do?
RTFM? Mind you, TFM is incredibly opaque.
The arguments that you give to x12() do not appear to be arguments of
that function.
Perhaps the usage has changed in the most recent update of the package.
Without
adequate warning.
E.g., there is, as far as I can discern, no "x12path" argument to
x12(). Whence it complains if
such an argument is supplied. When one loads the x12 package one gets a
message
saying that the path to the executables should be set using the function
x12path().
I would suggest that you try that. And then figure out how the other
arguments that you
supply should really be handled.
Perhaps try contacting the package maintainer. Good luck.
cheers,
Rolf Turner
More information about the R-help
mailing list