PAMA.Cov: This function implements Bayesian inference of PAMA model...

In PAMA: Rank Aggregation with Partition Mallows Model

Description

This function implements Bayesian inference of PAMA model with covariates.

Usage

PAMA.Cov(datfile, Covdatfile, nRe, iter = 1000, init = "EMM")

Arguments

datfile	A matrix or dataframe. This is the data where our algorithm will work on. Each colomn denotes a ranker's ranking. The data should be in entity-based format.
Covdatfile	A matrix or dataframe. Each column denotes a covariate.
nRe	A number. Number of relevant entities
iter	A number. Numner of iterations of MCMC. Defaulted as 1000.
init	A string. This indicates which method is used to initiate the starting point of the aggregated ranking list. "mean" uses the sample mean. "EMM" uses the method from R package 'ExtMallows'.

Details

The covariates are incoporated in the PAMA framework as indicators of group membership. That is, covariates are associated to group members via a logistic regression.

Value

List. It contains Bayesian posterior samples of all the parameters and log-likelihood.

1. I.mat: posterior samples of I

2. phi.mat: posterior samples of phi

3. smlgamma.mat: posterior samples of gamma

4. l.mat: posterior samples of log-likelihood.

5. theta.mat: posterior samples of coefficients of covariates.

Author(s)

Wanchuang Zhu, Yingkai Jiang, Jun S. Liu, Ke Deng

References

Wanchuang Zhu, Yingkai Jiang, Jun S. Liu, Ke Deng (2021) Partition-Mallows Model and Its Inference for Rank Aggregation. Journal of the American Statistical Association

Examples

a=NBANFL()
PAMA.Cov(t(a$NFLdata),a$NFLcov,nRe=10,iter=10)
## Not run: PAMA.Cov(t(a$NFLdata),a$NFLcov,nRe=10,iter=1000)

PAMA documentation built on May 6, 2021, 5:09 p.m.