Skip to content

evtimovr/missing_data_techniques_RI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

25 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Missing Data techniques for Reject Inference

Type: Master Thesis

Author: Radoslav Evtimov

1st Examiner: Prof. Dr. Stefan Lessmann

2nd Examiner: Prof. Dr. Benjamin Fabian

Idea behind the main output of the thesis:

results

Table of Content

Summary

Keywords: Reject Inference, Missing Data, Credit Risk, MNAR, Acceptance loop (give at least 5 keywords / phrases).

Full text: Full text can be found in different formats in the "paper" folder.

Working with the repo

Dependencies

For the results on the Master Thesis R 4.1.2 was used. There is dependence on the "miceMNAR" package that is currently not available on CRAN. It has to be manually downloaded and installed from https://cran.r-project.org/src/contrib/Archive/miceMNAR/.

Setup

  1. Clone this repository

  2. Set up your working directory in R to be the one with all files of this repo

Reproducing results

To recreate results, there are two options:

  • Run the predefined files for

    • 2 variables drawn from the Multivariate Gaussian Distribution Run files acceptance_loop_4.R to get the results corresponding to combination 4
    • 3 variables drawn from the Multivariate Gaussian Distribution Run file acceptance_loop_2.R to get the results corresponding to combination 2 Run file acceptance_loop_3.R to get the results corresponding to combination 3
  • Run the scripts in acceptance_loop_adjustables_4_vars.R and acceptance_loop_adjustables_5_vars.R for potentially different sets of combinations of parameters.

All dependecies will be automatically flagged - if any packages are missing from the environment, the console will return this as dependency. If they are available, they will be added to the environment.

Output: All files with results from the different combinations will be automatically stored in the working directory.

Results

The "plot" folder contains the plots shown in the paper. The actual results can be recreated by following the .R files in the "src" folder.

Project structure

├── README.md
├── paper
    ├── master_thesis_Evtimov_Radoslav.pdf          -- master Thesis in pdf format
    ├── master_thesis_Evtimov_Radoslav.tex          -- master Thesis in Latex format
    ├── references_master_thesis.bib                -- references in bibtex format
├── params                                          -- stores parameters  
├── plots                                           -- stores image file
└── src
    ├── acceptance_loop_2.R                         -- code for acceptance loop for combination 2
    ├── acceptance_loop_3.R                         -- code for acceptance loop for combination 3
    ├── acceptance_loop_4.R                         -- code for acceptance loop for combination 4
    ├── acceptance_loop_adjustables_4_vars.R        -- code for acceptance loop with adjustable combinations for 4 features
    ├── acceptance_loop_adjustables_5_vars.R        -- code for acceptance loop with adjustable combinations for 5 features
    ├── mnar_simulation_study.R                     -- code for MNAR simulation study
    ├── al_4_var_functions.R                        -- functions for the acceptance loop
    ├── al_5_var_functions.R                        -- functions for the acceptance loop
    ├── binaryPPMA_functions.R                      -- functions for the PPMM
    └── visualisations_master_thesis.R              -- visualisation script used                 

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published