legion

The package legion implements several multivariate models for purposes of forecasting.

Here is the list of the included functions:

1. ves - Vector Exponential Smoothing.
2. vets - Vector ETS with PIC taxonomy.
3. auto.vets - Automatic selection of restrictions for VETS.
4. sim.ves - simulates data from VES.
5. oves - occurrence state space vector exponential smoothing model.

Available methods:

1. AIC, BIC, AICc, BICc;
2. coefficients;
3. fitted;
4. forecast;
5. actuals;
6. logLik;
7. modelType - type of the estimated model;
8. nobs;
9. nparam - number of the estimated parameters in the model;
10. nvariate - number of series in the model;
11. residuals - the residuals of the model (et in case of additive and log(1+et) for the multiplicative ones);
12. rstandard, rstudent - standardised and studentised residuals;
13. outlierdummy - extracts outliers in the model and creates dummy variables for them;
14. plot - produces several plots for diagnostics purposes. See the documentation for plot.legion();
15. print;
16. sigma;
17. simulate;
18. summary;

Installation

A recent, development version, is available via github and can be installed using “devtools” in R. First, make sure that you have devtools: > if (!require(“devtools”)){install.packages(“devtools”)}

and after that run: > devtools::install_github(“config-i1/legion”)