Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market
Prices of non-ferrous metals have become more volatile due to rapid urbanisation and industrialisation (Liang et al., 2020, Wu and Hu, 2016). The objective of this paper is to assess risk indicators for non-ferrous metal market securities and to calculate an optimal portfolio. This paper uses a vari...
| Main Author: | |
|---|---|
| Format: | Dissertation (University of Nottingham only) |
| Language: | English |
| Published: |
2022
|
| Online Access: | https://eprints.nottingham.ac.uk/67901/ |
| _version_ | 1848800452354768896 |
|---|---|
| author | Li, Zhiyan |
| author_facet | Li, Zhiyan |
| author_sort | Li, Zhiyan |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Prices of non-ferrous metals have become more volatile due to rapid urbanisation and industrialisation (Liang et al., 2020, Wu and Hu, 2016). The objective of this paper is to assess risk indicators for non-ferrous metal market securities and to calculate an optimal portfolio. This paper uses a variety of methods to measure the risk of non-ferrous metal portfolios. The empirical results show that the historical simulation method, the age-weighted historical simulation method, and the Hull-White historical simulation method better measure the risk of a stock portfolio at a 95% confidence level. In addition, ES is a more accurate measure of risk than VaR when there are large shocks in non-ferrous markets. In conclusion, the age-weighted historical simulation method with a window of 1000 at 95% confidence level yields the best value of risk and passes the backtesting. The paper further uses mean-variance and mean-VaR models to solve for the optimal investment ratio for non-ferrous portfolios. The empirical results show that investors tend to invest in portfolio points on the efficient frontier. The tangent portfolio achieves optimal portfolio optimisation with the highest return/risk ratio. In addition, different models of stock returns, confidence levels and distribution fits all have an impact on the optimal portfolio. In summary, the application of the mean-VaR model under the assumption that returns follow a non-normal distribution yields better investment results than the mean-variance model and the mean-VaR model under the assumption that returns follow a normal distribution. |
| first_indexed | 2025-11-14T20:51:47Z |
| format | Dissertation (University of Nottingham only) |
| id | nottingham-67901 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:51:47Z |
| publishDate | 2022 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-679012023-04-25T15:48:41Z https://eprints.nottingham.ac.uk/67901/ Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market Li, Zhiyan Prices of non-ferrous metals have become more volatile due to rapid urbanisation and industrialisation (Liang et al., 2020, Wu and Hu, 2016). The objective of this paper is to assess risk indicators for non-ferrous metal market securities and to calculate an optimal portfolio. This paper uses a variety of methods to measure the risk of non-ferrous metal portfolios. The empirical results show that the historical simulation method, the age-weighted historical simulation method, and the Hull-White historical simulation method better measure the risk of a stock portfolio at a 95% confidence level. In addition, ES is a more accurate measure of risk than VaR when there are large shocks in non-ferrous markets. In conclusion, the age-weighted historical simulation method with a window of 1000 at 95% confidence level yields the best value of risk and passes the backtesting. The paper further uses mean-variance and mean-VaR models to solve for the optimal investment ratio for non-ferrous portfolios. The empirical results show that investors tend to invest in portfolio points on the efficient frontier. The tangent portfolio achieves optimal portfolio optimisation with the highest return/risk ratio. In addition, different models of stock returns, confidence levels and distribution fits all have an impact on the optimal portfolio. In summary, the application of the mean-VaR model under the assumption that returns follow a non-normal distribution yields better investment results than the mean-variance model and the mean-VaR model under the assumption that returns follow a normal distribution. 2022-02-10 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/67901/1/20260694_BUSI4019%20UNUK_2021.pdf Li, Zhiyan (2022) Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market. [Dissertation (University of Nottingham only)] |
| spellingShingle | Li, Zhiyan Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title | Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title_full | Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title_fullStr | Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title_full_unstemmed | Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title_short | Risk Metrics and Optimal Portfolio in Non-ferrous Metals Market |
| title_sort | risk metrics and optimal portfolio in non-ferrous metals market |
| url | https://eprints.nottingham.ac.uk/67901/ |