Chin Wen Cheong

Modelling financial observable-volatility using long memory models

This article proposes a generalized long persistence observable volatility model which comprises of leverage effect autoregressive fractionally integrated moving average model with time-varying volatility and the inclusion of heterogeneous autoregressive components as the contemporaneous variables. Our empirical results found that the proposed model provides substantial improvement in the model fitting as well as specification.

Adjusted Hurst exponent evaluations for equity and energy markets

Abstract This study introduced an alternative method in long memory volatility financial time series evaluation using the autoregressive conditional heteroscedasticity models. Instead of direct measurement for long memory Hurst exponent parameter, this approach suggested a short memory filtering procedures using the conditional heteroscedastic specification. The stationary time series which is free from short memory is later evaluated using two heuristic long memory estimations. It is found that this method is able to eliminate the possible spurious long memory in the selected equity and energy markets. The findings of this study are important in long memory estimation and market efficiency analysis.

Bivariate value-at-risk in the emerging Malaysian sectoral markets

Abstract This study examines the transmission of price changes and volatility among the Malaysian economic barometer (FTSE Bursa Malaysia Kuala Lumpur Composite Index- FBMKLCI) and four sectoral markets after the Asian financial crisis. The preliminary structural break identification provides an optimal sample size for the cross-markets transmission mechanism analysis. In order to reveal the hidden intention of interactions among the sectoral markets, the pair-wise markets which consist of CI-IND, CI-PLN, CI-PRP and CI-FIN are evaluated using a bivariate asymmetric BEKK model. The major intention of this study focuses on the cross-market hedging and market risk evaluations in terms of shocks and volatility.

The Stylized Facts and Market Risk Evaluations of Malaysian Shariah Index

Abstract This study investigates some interesting empirical stylized facts in the Malaysian Shariah index. These included the clustering volatility, leverage effect, fat-tailed behavior, long memory property and flexible power volatility representation. A conditional heteroscedasticity model is used to capture all the stylized facts. The stylized facts for Shariah index is compared with the major barometer indicators such as FTSE-KLCI, Singapore Strait-time index (STI) and Indonesian Jakarta Stock Exchange (JSE). The results of the analysis are used in quantifying the market risk using the value-at-risk approach.

The impact of subprime mortgage crisis to long-run and short-run volatility components of Indonesian and Malaysian equity markets

Abstract This study investigates the long-run and short-run movements of various stock market volatilities using a volatility decomposition methodology (Ding and Granger, 1996; Engle and Lee, 1999). We studied the impact of 2007–2008 subprime mortgage crisis on both the transitory and permanent volatility components in terms of two empirical stylized facts, the leverage effect and volatility persistence. In order to do so, the long spanning data are separated into three different periods. For the former stylized fact, the Crisis Impact on the leverage effect is mainly temporary with no long-run effect to the stock markets. This finding explained that the leverage effect is mostly difficult to adjust in the short-run transitory volatility during the crisis periods. However with proper risk management and long term strategies, most of the market participants are able to anticipate and handle this news impact in the long-run. For the latter stylized fact, the crisis has slightly increased the volatility persistence in all the markets. From the viewpoint of heterogeneous market hypothesis, the higher intensity of volatility persistence implied that the stock markets are less informationally efficient.

Measuring and forecasting Malaysian composite index volatility under the impact of Asian financial crisis - A revisit

Abstract This study revisited the impact of Asian Financial crisis to the Malaysian stock market. A generalized Tse’s model is used to depict various empirical stylized facts that occurred in the pre-and post-crisis periods. By utilizing the intraday volatility as the ex post of actual volatility, various forecasts horizons one-day ahead volatility are evaluated under six loss functions and regression analysis. It is found that the pre-crisis exhibited deeper impact of leverage effect, however less persistence volatility than the post-crisis period. The empirical results also shown that the asymmetric long memory model provided superior performance in the estimation as well as out-of-sample forecasts.

The value-at-risk evaluation of Brent’s crude oil market

This study investigates the market risk of the Brent’s crude oil market. First the long memory time-varying volatility is modelled under the Chung’s specification. Second, for model adequacy evaluations on the heavy-tailed, long memory and endogenously estimated power transformation models indicated superior performance in out-of-sample forecasts. Lastly, these findings are further applied in the long and short trading positions of market risk evaluations of the Brent’s market.

An empirical study of realized and long-memory GARCH standardized stock-return

In this article, we study the standardized returns by using the realized volatility and long-memory GARCH models. The various normality tests indicate that the realized–standardized returns follow a Gaussian distribution. On the other hand, the standardized returns by GARCH models are able to reduce but not eliminate the excess kurtosis condition compare to the realized–standardized returns.

The impact of subprime mortgage crisis on the short-run and long-run volatility components of the Malaysian stock market

This study investigates the long-run and short-run movements of two emerging stock market volatilities using a volatility decomposition methodology. We studied the impact of 2007–2008 subprime mortgage crisis on the transitory and permanent volatility components in terms of two empirical stylized facts, the leverage effect and volatility persistence. In order to do so, the long spanning data are separated into three different periods. For the former stylized fact, the crisis impact on the leverage effect is mainly temporary with no long-run effect to the stock markets. This finding explains that the leverage effect is mostly difficult to adjust in the short-run transitory volatility during the crisis periods. However with proper risk management and long term strategies, most of the market participants are able to anticipate and handle this news impact in the long-run. For the latter stylized fact, the crisis has slightly increased the volatility persistence in all the markets. From the viewpoint of heterogeneous market hypothesis, the higher intensity of volatility persistence implies the stock markets are less informational efficient.

Asymmetric and long memory volatility modelling for Asian equity markets

The fractionally integrated asymmetric power autoregressive conditional heteroscedasticity model has successfully captured the empirical stylized facts such as the leverage effect, volatility power transformation and long memory in the foreign exchange markets. This study further explores the applicability of this model in the Asian equity markets. The findings of this empirical study are important in understanding the underlying data generating processes and informationally efficient market analysis.