[R] Logistic Regression using MCMCpack

mcmc_logit.R

#!/usr/bin/env Rscript

sapply(c('dplyr', 'data.table', 'MASS', 'MCMCpack'), function(p) require(p, character.only = TRUE))

# Biopsy Data on Breast Cancer Patients
#
# variables:
#   'V1'     clump thickness.
#   'V4'     marginal adhesion.
#   'class'  "benign" or "malignant".
dt <- data.table(biopsy) %>% mutate(outcome = ifelse(class == 'malignant', 1, 0))

#
# Logistic Regression via MCMC
#
m <- MCMClogit(outcome ~ V1 + V4, data = dt, burnin = 1000, mcmc = 5000)

print(summary(m))
#
# Iterations = 1001:6000
# Thinning interval = 1
# Number of chains = 1
# Sample size per chain = 5000
#
# 1. Empirical mean and standard deviation for each variable,
#    plus standard error of the mean:
#
#                Mean      SD Naive SE Time-series SE
# (Intercept) -7.4592 0.58191 0.008229       0.025924
# V1           0.9207 0.08803 0.001245       0.003848
# V4           0.8182 0.08750 0.001237       0.004141
#
# 2. Quantiles for each variable:
#
#                2.5%     25%     50%     75%  97.5%
# (Intercept) -8.6666 -7.8185 -7.4571 -7.0591 -6.372
# V1           0.7567  0.8597  0.9173  0.9796  1.104
# V4           0.6570  0.7588  0.8136  0.8723  1.002
#