Name

vare — residuals from a VAR estimation

CALLING SEQUENCE

[resid]=vare(y,nlag,x)

PARAMETERS

Input

• y = an (nobs x neqs) matrix of y-vectors

• nlag = the lag length

• x = optional matrix of variables (nobs x nx)

• (NOTE: constant vector automatically included)

Output

• resid = matrix of residuals (nobs x neqs)

DESCRIPTION

Performs vector autogressive estimation and returns only residuals. Old function (but could be useful).

EXAMPLE


load(GROCERDIR+'/data/lutk1.dat')
bounds()
y=explone(['log(rfa_inv)';'log(rfa_inc)';'log(rfa_cons)')
trend=[1:size(y,1)]'
resid=vare(y,2,x)
 
// In this example, residuals from a VAR estimated with 2 lags, a trend and a constant as exogenous variables are calculated.

               

AUTHOR

Eric Dubois 2002