fillArows {crank} | R Documentation |
Imputes missing ranks using the Lim-Wolfe procedure
fillArows(x,maxcon=TRUE)
x |
A matrix of ranks that may contain ties and NAs. Columns represent objects ranked and rows represent ranking methods. |
maxcon |
Whether to impute rankings maximally consistent with the existing ones (TRUE) or minimally consistent (FALSE). |
fillArows
imputes missing ranks by examining the completed ranks for
each set of rows that have the same number of missing ranks. If more than one
row has the minimum number of missing values, the order of these rows is
permuted and the matrix x
becomes a list of matrices in which the
values in the rows will be imputed in different orders. Another level of
permutation and multiplication of matrices may occur in fillArow
to
which the matrices are passed for the actual imputation. The function
getLWargs
is called to get the arguments for fillArow
.
See Lim and Wolfe (2002) for details of this process.
A list of one or more completed matrices of ranks, possibly nested.
Jim Lemon
Lim, D.H. & Wolfe, D.A. (2002) An efficient alternative to average ranks for testing with incomplete ranking data. Biometrical Journal, 43(2): 187-206.
# The first example matrix from Lim and Wolfe (2002) lwmat<-matrix(c(3,1,2,4,NA,2,1,NA,2,NA,1,NA),nrow=3,byrow=TRUE) # complete with maximal consistency, permuting row order fillArows(lwmat) # now with minimal consistency as above fillArows(lwmat,maxcon=FALSE)