rank_sum_ratio: Rank Sum Ratio (RSR) Evaluation

In zhjx19/mathmodels: Comprehensive Mathematical Modeling Algorithms in R

Rank Sum Ratio (RSR) Evaluation

Description

Performs Rank Sum Ratio (RSR) evaluation on a dataset of positive indicators, computing ranks, weighted RSR values, and a linear regression model to fit RSR against probit-transformed ranks. Supports integer or non-integer ranking methods.

Usage

rank_sum_ratio(data, w = NULL, method = "int")

Arguments

data	Data frame with positive indicator data; first column is an ID column for identifying evaluation objects.
w	Numeric vector, weights for indicators (default = equal weights).
method	Character scalar, ranking method: "int" for integer ranks or "non-int" for scaled ranks in 1, n (default = "int").

Details

The rank_sum_ratio function implements the RSR method for evaluating objects based on positive indicators. It ranks the indicators (using integer or non-integer methods), computes weighted RSR values, adjusts ranks with probit transformation, and fits a linear regression model to relate RSR to probit values. The function assumes the first column of data is an ID column, and weights (w) can be provided or set to equal weights by default.

Value

A list containing:

• RSRtable: Data frame with RSR values, ranks, cumulative frequencies, probit values, and fitted RSR values.

• reg: Linear model object fitting RSR against probit values.

• rankTable: Data frame with ranked indicator values.

Examples

# Example data
data = data.frame(ID = c("A", "B", "C"), X1 = c(10, 20, 15), X2 = c(5, 10, 8))
w = c(0.4, 0.6)
rank_sum_ratio(data, w, method = "int")

zhjx19/mathmodels documentation built on June 2, 2025, 12:18 a.m.