-
If the trace=TRUE, this print the model and a value, what is this value ?, because its diferent for the ic values in the best model selected. In the example the value for the best model is 924.795
and…
pfv07 updated
9 years ago
-
Hi, I have the R version 3.2.0 in one computer and the version 3.2.2 in another and the same version of the package "forecast" (6.1) in both of them. So, when I apply the auto.arima function with d=1 …
-
Consider the following time series
```
require(forecast)
testseries
-
The auto.arima function has lots of functionality but does not have the same include.mean=F option that the Arima function has. Please can this be included as i'm currently having to modify the funct…
-
It would be nice to have a way to easily get the kind of stochastic process fitted. E.g. if we use auto.arima and we fit and ARIMA(2,1,1) with drift 0.6 it would be nice to have an easy way to get the…
-
Hi,
I have a similar problem as the issue with the title "auto.arima with Canova-Hansen test fails with too many zeroes on the series#31"
Here's my code:
library(forecast)
weight
-
Just three numbers and it fails to give a forecast, repro:
library(forecast ) # -> This is forecast 5.6
train = structure(c(1256.28753878144, 1217.71695906475, 1312.81301186437
), .Tsp = c(1, 3, 1),…
-
I know that probably it doesn't make sense to apply ARIMA in this timeseries I'll show you, but it is part of an automatic script in a server. Can you help me?
The error is: Error in if (mean(abs(reg…
-
Sometimes there are problems with finding a stable model for auto.arima and tbats. The response of the forecasting method then is:
Warning message:
In auto.arima(train) :
Unable to fit final model …
-
There is a new package in R called tsoutliers which detects outliers based on Chen and Liu method that apperared in the Journal of the American Statistical Association Volume 88, Issue 421, 1993.
For…