Mathematical models of copulas for assessing the risks of green projects
As the world embraces green technologies and projects, the need for precise modeling of complex interactions between distinct factors becomes increasingly important. Conventional standard modeling often assumes either independence or simple forms, such as Gaussian distributions, as key assumptions....
Збережено в:
| Дата: | 2025 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | Ukrainian |
| Опубліковано: |
Інститут проблем реєстрації інформації НАН України
2025
|
| Теми: | |
| Онлайн доступ: | http://drsp.ipri.kiev.ua/article/view/335614 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Data Recording, Storage & Processing |
Репозитарії
Data Recording, Storage & Processing| Резюме: | As the world embraces green technologies and projects, the need for precise modeling of complex interactions between distinct factors becomes increasingly important. Conventional standard modeling often assumes either independence or simple forms, such as Gaussian distributions, as key assumptions. However, these assumptions rarely become true in real-world scenarios.
For green risks, these approaches fail due to emergent behaviors that arise at the extremes of distributions, also known as tails. To address these challenges, this study proposes the use of copulas, which are well-regarded for their flexible and robust ability to model dependencies. A key focus of the research is on tail behavior, quantified using an approach similar to Value-at-Risk (VaR) estimation through empirical percentiles.
The calculations were performed using three different approaches: a naive approach assuming independent variables, a non-parametric method based on historical data, and a parametric method using copulas. The behavior of the dependencies was analyzed using data covering both economic and financial aspects of projects from various industries. Several copula families, including Gumbel, Clayton, and Frank, were tested, and their modeling performance was evaluated. The results demonstrated the ability of copulas to capture the dependencies accurately.
Compared to both naive and non-parametric methods, copulas showed notable advantages. They provided reliable extrapolations for data combinations that were not present in the original dataset, while effectively accounting for dependencies between factors. In conclusion, the importance of copulas in the green sector is expected to grow, leading to their increasing application as a replacement for simplistic models and resulting in more accurate risk assessments and decision-making in green projects. Tabl.: 2. Fig.: 1. Refs: 19 titles. |
|---|