| fillArow {crank} | R Documentation |
Impute a row of ranks using the existing values of rankings
Description
Imputes a row of missing ranks using the Lim-Wolfe procedure
Usage
fillArow(x,ranksums=NA,Arow,maxcon=TRUE)
Arguments
x |
A matrix of ranks that may contain ties and NAs. Columns represent objects ranked and rows represent ranking methods. |
ranksums |
The sums of ranks of all complete rows in ‘x’. |
Arow |
The row of ‘x’ that is to be completed. |
maxcon |
Whether to impute rankings maximally consistent with the existing ones (TRUE) or minimally consistent (FALSE). |
Details
‘fillArow’ imputes missing ranks in the row designated by ‘Arow’ using the information in ‘ranksums’. If the ranks already completed provide information on the order of imputation, that is used directly for imputed ranks of maximal consistency or inversely for imputed ranks of minimal consistency. If the existing ranks do not provide such information, the missing ranks are permuted, and a list of matrices with all the permutations is substituted. This may involve a recursive call to ‘fillArow’ and produce a nested list of matrices. See Lim and Wolfe (2002) for details of this process.
Value
The matrix ‘x’ with row ‘Arow’ completed or a list of such matrices, possibly nested.
Author(s)
Jim Lemon
References
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.
See Also
lwscreen, listBuilder, fillArows
Examples
# 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 the second row with maximal consistency
fillArow(lwmat,lwmat[1,],2)
# now with minimal consistency
fillArow(lwmat,lwmat[1,],2,maxcon=FALSE)