R/validnovcrit.R

Defines functions validnovcrit

Documented in validnovcrit

validnovcrit<-function(mat,criterion,H,r,p,tolval,tolsym){



################################################################
# validation of input that is specific to the criteria "TAU_2",#
#    "XI_2","ZETA_2", "CCR1_2" and "WALD"                      #    
################################################################

	if (!is.numeric(r)) stop("\n Argument r must be numeric.\n Perhaps unnamed arguments in the wrong order?\n")	
             
       	if (r<=0) stop("\n Argument r must be a positive integer (expected rank of effects matrix H)\n")
       	if (r!=floor(r)) {
                   warning("\n The specified expected rank 'r' of the effects matrix must be an integer. \n The specified value of the parameter 'r' has been truncated.\n")
                   r <- floor(r)
        }
     
	if (!is.matrix(H)) stop("\nEffect description data (H) is missing or is not given in matrix form \n")
	if (nrow(H) != p || ncol(H) != p) stop("\nEffect description matrix (H) does not have correct dimensions \n")
                
# checking symmetry of the H effects matrix
                
	maxabssym <- max(abs(H-t(H)))
	if (maxabssym > tolsym)
              {stop("\n The effect description matrix (H) supplied is not symmetric.\n Symmetric entries differ by up to ",maxabssym,".")}
      	else if (maxabssym > .Machine$double.eps) {
              H <-(H+t(H))/2
              warning("\n The effect description matrix (H) supplied was slightly asymmetric: \n symmetric entries differed by up to ",maxabssym,".\n (less than the 'tolsym' parameter).\n The H matrix has been replaced by its symmetric part.\n")
	}

# checking whether H matrix has the specified rank
                
	if (qr(H)$rank != r) {
                  if (criterion == "CCR1_2") stop("\nEffect description matrix (H) does not have the specified rank\n")
                  else warning("\n The expected rank of the effect matrix supplied by the user is ",r," but \n the matrix rank (as computed by R's qr function) is ",qr(H)$rank,".\n")
	}

# checking whether the E matrix is positive definite (not applicable to the Wald criterion)

                if (criterion != "WALD") {
			Eeigval = eigen(mat-H,only.values=T)$value

			if (Eeigval[p]/Eeigval[1] < -tolval) stop("\nError matrix (E=T-H) is not positive definite. Check whether the efects (H) matrix has been correctly specified.\n") 
		}

}

Try the subselect package in your browser

Any scripts or data that you put into this service are public.

subselect documentation built on July 26, 2023, 6:09 p.m.