Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange

It has been well documented that the empirical distribution of daily logarithmic returns from financial market variables is characterized by excess kurtosis and skewness. In order to capture such properties in financial data, heavy-tailed and asymmetric distributions are required to overcome shortfa...

Full description

Bibliographic Details
Main Authors: Huang, C., Huang, Chun-Kai, Chinhamu, K.
Format: Journal Article
Published: 2014
Online Access:http://hdl.handle.net/20.500.11937/68211
_version_ 1848761772489572352
author Huang, C.
Huang, Chun-Kai
Chinhamu, K.
author_facet Huang, C.
Huang, Chun-Kai
Chinhamu, K.
author_sort Huang, C.
building Curtin Institutional Repository
collection Online Access
description It has been well documented that the empirical distribution of daily logarithmic returns from financial market variables is characterized by excess kurtosis and skewness. In order to capture such properties in financial data, heavy-tailed and asymmetric distributions are required to overcome shortfalls of the widely exhausted classical normality assumption. In the context of financial forecasting and risk management, the accuracy in modeling the underlying returns distribution plays a vital role. For example, risk management tools such as value-at-risk (VaR) are highly dependent on the underlying distributional assumption, with particular focus being placed at the extreme tails. Hence, identifying a distribution that best captures all aspects of the given financial data may provide vast advantages to both investors and risk managers. In this paper, we investigate major financial indices on the Johannesburg Stock Exchange (JSE) and fit their associated returns to classes of heavy tailed distributions. The relative adequacy and goodness-of-fit of these distributions are then assessed through the robustness of their respective VaR estimates. Our results indicate that the best model selection is not only variant across the indices, but also across different VaR levels and the dissimilar tails of return series.
first_indexed 2025-11-14T10:36:59Z
format Journal Article
id curtin-20.500.11937-68211
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T10:36:59Z
publishDate 2014
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-682112018-05-18T08:01:20Z Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange Huang, C. Huang, Chun-Kai Chinhamu, K. It has been well documented that the empirical distribution of daily logarithmic returns from financial market variables is characterized by excess kurtosis and skewness. In order to capture such properties in financial data, heavy-tailed and asymmetric distributions are required to overcome shortfalls of the widely exhausted classical normality assumption. In the context of financial forecasting and risk management, the accuracy in modeling the underlying returns distribution plays a vital role. For example, risk management tools such as value-at-risk (VaR) are highly dependent on the underlying distributional assumption, with particular focus being placed at the extreme tails. Hence, identifying a distribution that best captures all aspects of the given financial data may provide vast advantages to both investors and risk managers. In this paper, we investigate major financial indices on the Johannesburg Stock Exchange (JSE) and fit their associated returns to classes of heavy tailed distributions. The relative adequacy and goodness-of-fit of these distributions are then assessed through the robustness of their respective VaR estimates. Our results indicate that the best model selection is not only variant across the indices, but also across different VaR levels and the dissimilar tails of return series. 2014 Journal Article http://hdl.handle.net/20.500.11937/68211 restricted
spellingShingle Huang, C.
Huang, Chun-Kai
Chinhamu, K.
Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title_full Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title_fullStr Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title_full_unstemmed Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title_short Assessing the relative performance of heavy-tailed distributions: Empirical evidence from the Johannesburg stock exchange
title_sort assessing the relative performance of heavy-tailed distributions: empirical evidence from the johannesburg stock exchange
url http://hdl.handle.net/20.500.11937/68211