Aggregating results from Arima runs R

Question

/edit: To clarify: The mtable function from the memisc package does exactly what I need, but unfortunately does not work with arima models.

Similar to this question: I have multiple Arima models, some of which I've also fit with dependent variables. I'd like an easy way to make a table/graph of the coefficients in each model, as well as summary statistics about each model.

Here is some example code:

sim <- arima.sim(list(order = c(1,1,0), ar = 0.7), n = 200)

ar1<-arima(sim,order=c(1,1,0))
ar2<-arima(sim,order=c(2,1,0))
ar3<-arima(sim,order=c(3,1,0))
ar4<-arima(sim,order=c(2,2,1))

#Try mtable
library(memisc)
mtable("Model 1"=ar1,"Model 2"=ar2,"Model 3"=ar3,"Model 4"=ar4)
#>Error in UseMethod("getSummary") : 
#  no applicable method for 'getSummary' applied to an object of class "Arima"

#Try  apsrtable
library(apsrtable)
apsrtable("Model 1"=ar1,"Model 2"=ar2,"Model 3"=ar3,"Model 4"=ar4)
#>Error in est/x$se : non-numeric argument to binary operator

Answer 1

You need to write a function called getSummary.Arima for use with the memisc package. Check out getSummary.lm for an example.