 Usage:  {B,u,s,g} = varunls(y,p,trend)  
 
 Input:

  y                      (K x p+T) matrix of p presample, and T sample observations 
                         
  p                      (1 x 1) integer, VAR-process order, number of lags in the model, p=1,2,3,... 
                         
  trend                  (1 x 1) integer, indicator whether intercept is estimated or not, if trend=1 intercept is estimated, if trend=0 no intercept is estimated 
                         
 Output:

  B                      (K x trend+K*p) matrix, model parameters (nu~)A_1~A_2~...~A_p 
                         
  u                      (K x T) matrix, least squares estimates of residuals 
                         
  s                      (K x K) matrix, least squares estimate of residual variance-covariance matrix 
                         
  g                      (K*p+trend x K*p+trend) matrix, autocovariance matrix of time series 
                         
--------------------------------------------------------------
(C) MD*TECH Method and Data Technologies, 17.8.2000
