This application generates some types ARMA time series models according to described in the book "Time series analysis, forecasting and control" from Box & Jenkins. The inputs required by the user are:
- φ1 for autoregressive time series AR(1);
- φ1 and φ2 for autoregressive time series AR(2);
- θ1 for moving average time series MA(1);
- θ1 and θ2 for moving average time series MA(2);
- φ1 and θ1 for autoregressive/moving average time series ARMA(1,1);
- Sample size, minimum value is 50 and maximum value is 250;
- Mean of the time series;
- at() noise variance, value must be greater than zero.
External libs required:
- GUI - MigLayout https://www.miglayout.com/
- Charts - JFreChart https://www.jfree.org/jfreechart/
- Export excel tables - Apache POI https://poi.apache.org/
For your convenience there is an runnable jar file in the runnable_jar directory. Tested using Windows 10 Home Single Language 21H2.
Built using Java S.E. 1.7
Some application screenshots:
**MAIN WINDOW**
**CALCULATION RESULTS**
**GRAPH**
**HELP WINDOW**
