j0.multi {TopKLists}

Function returning an overall point j0 of degeneration into noise for multiple ranked lists

Description

Moderate deviation-based calculation of an overall point j_0 of degeneration into noise for multiple ranked lists. The function takes a matrix of ordered lists and estimates a j_0 for each pair of the input lists (columns), with repect to the preselected distance parameter \delta. This function combines the functions compute.stream and prepare.Idata.

Usage

j0.multi(lists, d, v)

Arguments

lists

Input data frame, where each column represents one list of ordered items

d

The maximal distance of an object's rank positions when two lists are compared. When the distance between the respective rank positions of the object is smaller or equal d, then the object is assigned the value 1, otherwise 0

v

Parameter for estimating j_0

Details

The smaller d, the stronger the assumption about the concordance of any two lists (d=0 is assuming identical rankings of an object)

Value

A list containing the maximal estimated indices of information degradation j_0 through all combinations of L lists:

maxK

Maximal estimated k through all combinations of two lists

L

Data frame of estimated j_0 for each pairwise comparison

Idata

Data stream vector of zeros and ones

Author(s)

Eva Budinska <[e-mail]>, Michael G. Schimek <[e-mail]>

References

Hall, P. and Schimek, M. G. (2012). Moderate deviation-based inference for random degeneration in paired rank lists. J. Amer. Statist. Assoc., 107, 661-672.

See Also

compute.stream, prepare.idata

Examples

set.seed(4657)

lists <- data.frame(L1=c("A","B","C","D","E","F","G","H","J","I","K","L","M","N"))
lists$L2 <- c("B","C","A","E","G","F","G","J","K","L","M","N","I","H")
lists$L3 <- sample(LETTERS[1:14])
res.j0.temp = j0.multi(lists, d=5, v=3)