CODEWITHSUNDEEP
rarimaforecast
MayankT
MayankT

Reputation: 19

Interpreting plot(auto.arima) in R

I was playing with the package forecast and created the model: 

mod <- auto.arima(univariate time series)

and tried running 

plot(mod)

enter image description here

I am interested in knowing how to interpret this result.

Upvotes: 1

Views: 2668

Answers (1)

Rafael D&#237;az
Rafael D&#237;az

Reputation: 2289

When modelling data with ARIMA models, it is sometimes useful to plot the inverse characteristic roots. Plot auto.arima function will compute and plot the inverse roots for any fitted ARIMA model (including seasonal models). The plot return the autoregressive roots from the AR characteristic polynomial and return the moving average roots from the MA characteristic polynomial.

The plot(mod) function will plot the inverse of the roots on the complex unit circle. A causal invertible model should have all the roots outside the unit circle. Equivalently, the inverse roots should like inside the unit circle. If all roots have modulus less than one and lie inside the unit circle, then the estimated ARMA is stable (stationary) and invertible and therefore will give good estimates.

Upvotes: 4

Related Questions