forecasts from a VARMA model
rvarmaf = varmaf(namevarma,hfore,arg1,...,argn)
* namevarma = the tlist results resulting from a VARMA estimation or the same between quotes (to have its name saved)
* hfore = the prevision period which can be either
-a [n1 n2] constant vector where n1 and n2 are the lead over the last period of the estimation (n1<=0 means that the forecast begins within the estimation period)
-a n constant which is equivalent to [1 n] (forecast begins just after the estimation period)
-a [n1 n2] string vector where n1 and n2 are the time periods for forecasting (a posibility open only if the VAR has been estimated on ts)
-a n string which is equivalent to [1 n] (forecast begins just after the estimation period)
* argi = a variables # of arguments that can be:
- 'exo=xx' (necessary if the variables in the VARMA has been given as a matrix) where xx is either a vector of strings betweeen double quotes, a list of ts or a matrix of size [# of forecasting dates ; # of exogenous variables in the var] for the values of exogenous variables over the forecasting period
- the string 'noprint' if the user doesn't want to print the results of the forecast
* rvarmaf = a results tlist with:
- result('meth') ='varmaf'
- rvarmaf('namevarma') = the name of the varma tlist result (if entered between quotes)
- rvarmaf('rvarma') = the varma tlist result used for forecasting
- rvarmaf('h forecast') = the forecast horizon
- rvarmaf('forecast') = the forecast
- rvarmaf('forecast var') = the forecast variance
- rvarmaf('fore_namei') = vector or ts of forecast called by the name of the forecast variable prefixed by 'fore_'
load(GROCERDIR+'/data/bdhenderic.dat') ; bounds('1964q3','1988q4'); rarma=varma('rnet',[0 0],[],[0],[],0,1,'exo=''const''','Gexo=1') // estimation of an ARMA(2,1) on the interest data in the database bdhenderic.dat provided with grocer rarmf=varmaf(rarma,10,'exo=''const''') // makes a forecast on the first 10 out-of-sample observations rarmf=varmaf(rarma,'1991q2','exo=''const''') // makes a forecast from the first out-of-sample date (here '1989q1') to '1991q2': this is exactly the same as the first example rarmf=varmaf(rarma,['1988q3' '1990q4'],'exo=''const''') // makes a forecast from an in-sample date ('1988q3') to an out-of-sample date ('1990q4') rarmf=varmaf(rarma,['1989q3' '1990q4'],'exo=''const''') // makes a forecast from a date three quarters after the end of estimation period to an out-of-sample date; // note that it is possible since data on rnet are available until 1989q4, // that is after the start of the forecast. | ![]() | ![]() |