---
title: XGBoost for LGD Approximation
url: https://www.ml-quant.com/papers/repec/war-wpaper-2024-12/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: RePEc:war:wpaper:2024-12
source_url: https://econpapers.repec.org/scripts/redir.pf?u=https%3A%2F%2Fwww.wne.uw.edu.pl%2Fdownload_file%2F4362%2F0%3Bh%3Drepec%3Awar%3Awpaper%3A2024-12
featured: 2024-07-03
citations: unknown
topic: ML & AI Methods
---


# XGBoost for LGD Approximation

The study uses machine learning to enhance the accuracy of Loss Given Default (LGD) estimation in situations with limited cash-flow data, using a European mortgage portfolio.

- Source: https://econpapers.repec.org/scripts/redir.pf?u=https%3A%2F%2Fwww.wne.uw.edu.pl%2Fdownload_file%2F4362%2F0%3Bh%3Drepec%3Awar%3Awpaper%3A2024-12
- Identifier: RePEc:war:wpaper:2024-12
- Released: 2024-07-03
- First featured: Quant Letter No. 55 (2024-07-03): https://www.ml-quant.com/issues/2024-07-03/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: ML & AI Methods

## Related

- [Real Estate Appraisals: ML vs Traditional Methods](https://www.ml-quant.com/papers/repec/spr-gjorer-v-9-y-2023-i-2-d-10-1365-s41056-022-00063-1/): ML vs Traditional Methods: Research indicates that XGBoost, a machine learning method, offers the most precise estimates in automated property valuations, suggesting a need for regulators to use multiple methods.
- [Wasserstein Gradient Boosting: A Framework for Distribution-Valued Supervised Learning](https://www.ml-quant.com/papers/arxiv/2405.09536/): Wasserstein gradient boosting, a new type of gradient boosting, improves probabilistic prediction by approximating the output-distribution parameter's posterior distribution.
- [Using Images as Covariates: Measuring Curb Appeal with Deep Learning](https://www.ml-quant.com/papers/arxiv/2403.19915/): Incorporating image data into econometric models through deep learning enhances the accuracy of residential real estate price predictions.
- [Financial Fraud Detection System Based on Improved Random Forest and Gradient Boosting Machine (GBM)](https://www.ml-quant.com/papers/arxiv/2502.15822/): The paper suggests a financial fraud detection system that uses an improved Random Forest and Gradient Boosting Machine model, offering an efficient and reliable solution for detecting financial fraud.
- [AI Realtor: Towards Grounded Persuasive Language Generation for Automated Copywriting](https://www.ml-quant.com/papers/arxiv/2502.16810/): The paper introduces a system that uses large language models to create persuasive, personalized real estate marketing content, which was favored over human-written descriptions in tests, indicating potential for automated, fact-based targeted marketing.
- [Stock Price Crash Prediction Based on Multimodal Data Machine Learning Models](https://www.ml-quant.com/papers/ssrn/4575784/): The paper suggests a machine learning framework that predicts stock market crashes by combining market data, graph data, and sentiment analysis, with LightGBM showing superior accuracy.