Return to Question

Not sure if this is a bug or a typo but for the first example of the ARMAProcess in Mathematica 9 we have:

data = Select[WeatherData["Champaign", "Temperature", {2012, 8}], FreeQ[#, Missing] &][[All, 2]];
temp = TemporalData[data]
eproc = EstimatedProcess[temp, ARMAProcess[3, 3]]

EstimatedProcess hangs, whereas if I just send the data (with RLink!) to R's arima solver (with order=c(3,0,3) so it is an ARMA model), it solves instantly, and all further analysis looks the same as the example, so it isn't given a bad solution or anything.

Is there something wrong with the documentation (reminds me of the issues of trying to do optimization without using ?NumericQ variables).

Any thoughts? Really want to use this feature...

Not sure if this is a bug or a typo but for the first example of the ARMAProcess in Mathematica 9 we have:

data = Select[WeatherData["Champaign", "Temperature", {2012, 8}], 
              FreeQ[#, Missing] &][[All, 2]];
temp = TemporalData[data]
eproc = EstimatedProcess[temp, ARMAProcess[3, 3]]

EstimatedProcess hangs, whereas if I just send the data (with RLink!) to R's arima solver (with order=c(3,0,3) so it is an ARMA model), it solves instantly, and all further analysis looks the same as the example, so it isn't given a bad solution or anything.

Is there something wrong with the documentation (reminds me of the issues of trying to do optimization without using ?NumericQ variables).

Any thoughts? Really want to use this feature...