Regional Characteristics And Anomaly Detection Of KODIT Default Rates Using Prophet

Background: Small and medium-sized enterprises (SMEs) are the backbone of the national economy, necessitating a stable financial environment. This study focuses on analyzing default rates from Korea Credit Guarantee Fund (KODIT) data across 18 regions in South Korea and detecting anomalies using the Prophet model. Prophet is recognized as a powerful tool for time series analysis, capable of effectively separating trends, seasonality, and external factors while identifying anomalies. Materials and Methods: This study utilized KODIT default rate data spanning from 2012 to 2023. The data were preprocessed to fit the Prophet model's requirements, and information on national holidays and weekends was incorporated to enhance anomaly detection performance. The preprocessing ensured the model accurately captured temporal patterns and external influences within the data. Results: The analysis revealed a common pattern of increasing default rates across most regions after 2022, likely reflecting the impact of external factors such as the COVID-19 pandemic. Notably, Sejong and Jeju exhibited higher frequencies and intensities of anomalies compared to other regions, highlighting the need for further exploration of the relationship between regional economic characteristics and external factors. Furthermore, the Prophet model demonstrated high predictive accuracy in economically stable regions like Seoul and Gyeonggi, but its anomaly detection performance was relatively limited in regions like Gyeongnam and Ulsan, dominated by manufacturing, and Jeonbuk and Jeonnam, centered around agriculture. Conclusion: This study demonstrates that the Prophet model is a valuable tool for detecting anomalies and devising credit risk management strategies. While it underscores the importance of exploring quantitative correlations between anomalies and external factors, it also suggests that Prophet-based analyses can extend beyond credit risk management to broader economic decision-making processes

Authors

Publication Details

Journal
IOSR Journal of Computer Engineering
Published
2024-12-01
DOI
https://doi.org/10.9790/0661-2606020917
Primary Topic
Technology and Data Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Regional Characteristics And Anomaly Detection Of KODIT Default Rates Using Prophet

Juryon Paik
IOSR Journal of Computer Engineering
Technology and Data Analysis
article

Regional Characteristics And Anomaly Detection Of KODIT Default Rates Using Prophet

Juryon Paik
article en

Abstract

Background: Small and medium-sized enterprises (SMEs) are the backbone of the national economy, necessitating a stable financial environment. This study focuses on analyzing default rates from Korea Credit Guarantee Fund (KODIT) data across 18 regions in South Korea and detecting anomalies using the Prophet model. Prophet is recognized as a powerful tool for time series analysis, capable of effectively separating trends, seasonality, and external factors while identifying anomalies. Materials and Methods: This study utilized KODIT default rate data spanning from 2012 to 2023. The data were preprocessed to fit the Prophet model's requirements, and information on national holidays and weekends was incorporated to enhance anomaly detection performance. The preprocessing ensured the model accurately captured temporal patterns and external influences within the data. Results: The analysis revealed a common pattern of increasing default rates across most regions after 2022, likely reflecting the impact of external factors such as the COVID-19 pandemic. Notably, Sejong and Jeju exhibited higher frequencies and intensities of anomalies compared to other regions, highlighting the need for further exploration of the relationship between regional economic characteristics and external factors. Furthermore, the Prophet model demonstrated high predictive accuracy in economically stable regions like Seoul and Gyeonggi, but its anomaly detection performance was relatively limited in regions like Gyeongnam and Ulsan, dominated by manufacturing, and Jeonbuk and Jeonnam, centered around agriculture. Conclusion: This study demonstrates that the Prophet model is a valuable tool for detecting anomalies and devising credit risk management strategies. While it underscores the importance of exploring quantitative correlations between anomalies and external factors, it also suggests that Prophet-based analyses can extend beyond credit risk management to broader economic decision-making processes

IOSR Journal of Computer EngineeringVol. 26(6)
Openalex Percentile: Top 65%
Technology and Data Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Regional Characteristics And Anomaly Detection Of KODIT Default Rates Using Prophet — Juryon Paik · IOSR Journal of Computer Engineering (2024) | TGRS Research Map | TGRS