Estimation.Zemel.MLE: Estimates Zemel Parameters via Gradient Descent

In StatRank: Statistical Rank Aggregation: Inference, Evaluation, and Visualization

Description

This function takes in data broken into pairs, and estimates the parameters of the Zemel mode via Gradient Descent

Usage

Estimation.Zemel.MLE(Data.pairs, m, threshold = 1e-04,
  learning.rate = 1/30000)

Arguments

Data.pairs	data broken up into pairwise comparisons
m	how many alternatives
threshold	turning parameter for gradient descent
learning.rate	turning parameter for gradient descent

Value

a set of scores for the alternatives, normalized such that the sum of the log scores is 0 scores <- Generate.Zemel.Parameters(10)$Score pairs <- Generate.Zemel.Ranks.Pairs(scores, 10, 10) Estimation.Zemel.MLE(pairs, 10, threshold = .1)

StatRank documentation built on May 1, 2019, 8:22 p.m.