Hi,
my name is Jaromir Baxa and I have problem with ARIMA modelling using fSeries package. This is what I did (same thing as last week with another data) and what R responded me:
res <- read.table("E:/BusinessCyclesTheory/SAPres.txt")
res <- edit(res)
library(fSeries)
Loading required package: fBasics
Rmetrics, (C) 1999-2005, Diethelm Wuertz, GPL
fBasics: Markets, Basic Statistics, Date and Time
Loading required package: fCalendar
Rmetrics, (C) 1999-2005, Diethelm Wuertz, GPL
fCalendar: Markets, Basic Statistics, Date and Time
Rmetrics, (C) 1999-2005, Diethelm Wuertz, GPL
fSeries: The Dynamical Process Behind Financial Markets
armaFit(formula = res ~ arima(1, 0, 1))
Error in fit$x - fit$residuals : non-numeric argument to binary operator
In addition: Warning message:
Incompatible methods ("Ops.data.frame", "Ops.ts") for "-"
"res" is vector of residuals (1 column, 119 observations) obtained from Henderson moving average.
Does anybody know what this error does mean and how can I avoid it (by transforming data??)?
Thanks for helping me. JB.
| Hi,
| my name is Jaromir Baxa and I have problem with ARIMA modelling using fSeries package. This is what I did (same thing as last week with another data) and what R responded me:
| > res <- read.table("E:/BusinessCyclesTheory/SAPres.txt")
| > res <- edit(res)
| > library(fSeries)
[...]
| > armaFit(formula = res ~ arima(1, 0, 1))
| Error in fit$x - fit$residuals : non-numeric argument to binary operator
| In addition: Warning message:
| Incompatible methods ("Ops.data.frame", "Ops.ts") for "-"
|
| "res" is vector of residuals (1 column, 119 observations) obtained from Henderson moving average.
| Does anybody know what this error does mean and how can I avoid it (by
| transforming data??)?
I would try to narrow it down by using the ARIMA command directly --- this is
what fSeries calls anyway, and there may just be a bug in the wrapping which
you'd avoid. See help(arima).
Hope this helps, Dirk
Statistics: The (futile) attempt to offer certainty about uncertainty.
-- Roger Koenker, 'Dictionary of Received Ideas of Statistics'