Learn R Programming

COMBO

COMBO: COrrecting Misclassified Binary Outcomes

Overview

COMBO provides a set of functions for the analysis of regression models with binary outcome misclassification.

The two main parts are:

Classification probability calculations
Parameter estimation

Classification probability calculations

The package allows users to compute the probability of the latent true outcome and the conditional probability of observing an outcome given the latent true outcome, based on parameters estimated from the COMBO_EM and COMBO_MCMC functions.

Parameter estimation

Jointly estimate parameters from the true outcome and observation mechanisms, respectively, in a binary outcome misclassification model using the EM algorithm or MCMC. Parameters from the true outcome, first-stage observation, and second-stage observation mechanisms in a two-stage binary outcome misclassification model can also be estimated using the EM algorithm and MCMC.

Installation

# Install from CRAN
install.packages("COMBO")

# Install the development version from GitHub:
# install.packages("devtools")
devtools::install_github("kimberlywebb/COMBO")

Please note that COMBO requires JAGS to be installed. JAGS can be downloaded from https://sourceforge.net/projects/mcmc-jags/.

Copy Link

Version

Install

install.packages('COMBO')

Monthly Downloads

545

Version

1.1.0

License

MIT + file LICENSE

Maintainer

Kimberly Hochstedler

Last Published

July 6th, 2024

Functions in COMBO (1.1.0)

mean_pistarjj_compute

Compute the Mean Conditional Probability of Correct Classification, by True Outcome Across all Subjects

Expected Complete Data Log-Likelihood Function for Estimation of the Two-Stage Misclassification Model

Fix Label Switching in MCMC Results from a Binary Outcome Misclassification Model

EM-Algorithm Function for Estimation of the Misclassification Model

Expected Complete Data Log-Likelihood Function for Estimation of the Misclassification Model

jags_picker_2stage

Set up a Two-Stage Binary Outcome Misclassification jags.model Object for a Given Prior

label_switch_2stage

Fix Label Switching in MCMC Results from a Binary Outcome Misclassification Model

Set up a Binary Outcome Misclassification jags.model Object for a Given Prior

em_function_2stage

EM-Algorithm Function for Estimation of the Two-Stage Misclassification Model

naive_jags_picker_2stage

Set up a Naive Two-Stage Regression jags.model Object for a Given Prior

misclassification_prob

Compute Conditional Probability of Each Observed Outcome Given Each True Outcome, for Every Subject

naive_jags_picker

Set up a Naive Logistic Regression jags.model Object for a Given Prior

Select a Binary Outcome Misclassification Model for a Given Prior

misclassification_prob2

Compute Conditional Probability of Each Second-Stage Observed Outcome Given Each True Outcome and First-Stage Observed Outcome, for Every Subject

naive_model_picker_2stage

Select a Naive Two-Stage Regression Model for a Given Prior

perfect_sensitivity_EM

EM-Algorithm Estimation of the Binary Outcome Misclassification Model while Assuming Perfect Sensitivity

model_picker_2stage

Select a Two-Stage Binary Outcome Misclassification Model for a Given Prior

pitilde_by_chain

Compute the Mean Conditional Probability of Second-Stage Correct Classification, by First-Stage and True Outcome Across all Subjects for each MCMC Chain

naive_loglik_2stage

Observed Data Log-Likelihood Function for Estimation of the Naive Two-Stage Misclassification Model

naive_model_picker

Select a Logisitic Regression Model for a Given Prior

pitilde_compute

Compute Conditional Probability of Each Second-Stage Observed Outcome Given Each True Outcome and First-Stage Observed Outcome, for Every Subject

M-Step Expected Log-Likelihood with respect to Beta

pitilde_compute_for_chains

Compute Conditional Probability of Each Observed Outcome Given Each True Outcome for a given MCMC Chain, for Every Subject

Sum Every "n"th Element

Sum Every "n"th Element, then add 1

pistar_by_chain_2stage

Compute the Mean Conditional Probability of Correct Classification, by True Outcome Across all Subjects for each MCMC Chain for a 2-stage model

Compute E-step for Two-Stage Binary Outcome Misclassification Model Estimated With the EM-Algorithm

Compute Conditional Probability of Each Observed Outcome Given Each True Outcome, for Every Subject

Compute Probability of Each True Outcome, for Every Subject

pistar_by_chain

Compute the Mean Conditional Probability of Correct Classification, by True Outcome Across all Subjects for each MCMC Chain

M-Step Expected Log-Likelihood with respect to Gamma

M-Step Expected Log-Likelihood with respect to Delta

pistar_compute_for_chains_2stage

Compute Conditional Probability of Each Observed Outcome Given Each True Outcome for a given MCMC Chain, for Every Subject for 2-stage models

Compute E-step for Binary Outcome Misclassification Model Estimated With the EM-Algorithm

pistar_compute_for_chains

Compute Conditional Probability of Each Observed Outcome Given Each True Outcome for a given MCMC Chain, for Every Subject

true_classification_prob

Compute Probability of Each True Outcome, for Every Subject

Example data from The Law School Admissions Council's (LSAC) National Bar Passage Study (Linda Wightman, 1998)

MCMC Estimation of the Binary Outcome Misclassification Model

EM-Algorithm Estimation of the Binary Outcome Misclassification Model

COMBO_EM_2stage

EM-Algorithm Estimation of the Two-Stage Binary Outcome Misclassification Model

check_and_fix_chains_2stage

Check Assumption and Fix Label Switching if Assumption is Broken for a List of MCMC Samples

COMBO_MCMC_2stage

MCMC Estimation of the Two-Stage Binary Outcome Misclassification Model

COMBO_data_2stage

Generate data to use in two-stage COMBO Functions

check_and_fix_chains

Check Assumption and Fix Label Switching if Assumption is Broken for a List of MCMC Samples

Generate Data to use in COMBO Functions

Test data for the COMBO_EM function