Understanding ARIMA

GIS

New Member
#1
Hi,
I am trying to decompose time series contains approx. 120 observations.
Using ARIMA(p,d,q) model I need to find three parameters:
d - this one I found, by differencing once the observations (hence the mean and variance have turned pretty constant over time).
As to the other two: p, q - I have been using R's functions ACF and PACF in order to examine autocorrelation and partial autocorrelation, respectively.
In the attach files you can see the both plots - where the x axis represents number of lags (I chose 10 lags).
Could anyone help with identifying what should I use for p and q?
Hope my question is clear,
Cheers