https://github.com/seungwoo-stat/20th-pe-forecast

Forecasting the four-party vote share of the 20th presidential election of Republic of Korea
https://github.com/seungwoo-stat/20th-pe-forecast

compositional-data presidential-election

Last synced: 12 days ago
JSON representation

Forecasting the four-party vote share of the 20th presidential election of Republic of Korea

• Host: GitHub
• URL: https://github.com/seungwoo-stat/20th-pe-forecast
• Owner: seungwoo-stat
• License: mit
• Created: 2022-05-06T05:44:08.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2023-11-23T07:32:49.000Z (over 2 years ago)
• Last Synced: 2025-10-30T03:13:21.874Z (9 months ago)
• Topics: compositional-data, presidential-election
• Language: R
• Homepage:
• Size: 16.8 MB
• Stars: 1
• Watchers: 1
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# 20th Presidential Election Forecast

Forecasting the four-party vote share of the 20th presidential election of the Republic of Korea

![](nat-trend-small.png)

## Overview

-   Execute the `R` files in each folder in the order of the folders' name: `1a_data_generate` - `1b_summary` - `2_fundamental_model` - `3_poll_model`

-   `1a_data_generate` folder

    -   `save_RData.R` executes all three `R` files in the same folder. By result, it generates `pe.RData` file that saves historical presidential election results and 20th presidential election pre-election polls.

-   `1b_summary` folder

    -   Use `basic_plot.R` to plot summary of the historical two-party vote share of the Democratic Party. This plot is included in our manuscript as figure 1.

-   `2_fundamental_model` folder

    -   Run `run-fund-model.R` to conduct posterior sampling using Rstan. Rstan codes are in `fund-model-simple.stan` file.

    -   Use `fund-model-plot.R` to make summary plots from the posterior samples.

-   `3_poll_model` folder

    -   Run `run-poll-model.R` to conduct posterior sampling using Rstan. Rstan codes are in `poll-model.stan` file.

    -   Use `poll-model-plot.R` to make summary plots from the posterior samples.

## Reference

-   Seungwoo Kang and Hee-Seok Oh. (2024) [Forecasting South Korea's Presidential Election via Multiparty Dynamic Bayesian Modeling](https://doi.org/10.1016/j.ijforecast.2023.01.004). *International Journal of Forecasting*. **40**(1), pp. 124-141.