Displaying 1 result from an estimated 1 matches for "0.104707".
Did you mean:
0.104007
2011 Jul 04
1
forecast: bias in sampling from seasonal Arima model?
Dear all,
I stumbled upon what appears to be a troublesome issue when sampling from an
ARIMA model (from Rob Hyndman's excellent 'forecast' package) that contains
a seasonal AR component.
Here's how to reproduce the issue. (I'm using R 2.9.2 with forecast 2.19;
see sessionInfo() below).
First some data:
> x <- c(
0.132475, 0.143119, 0.108104, 0.247291, 0.029510,