Assoc. Prof. Dr. Ng Kok Haur
Associate Professor
Institute of Mathematical Sciences
Faculty of Science
Universiti Malaya
kokhaur@um.edu.my

NO	NAME	INSTITUTION	FACULTY/SCHOOL/ CENTRE/UNIT
1	Assoc. Prof. Dr. Mahatelge Shelton Peiris	University of Sydney	Faculty of Science
2	Prof. Dr. Ibrahim bin Mohamed	UM	Faculty of Science
3	Dr. Ng Kooi Huat	UTAR	Faculty of Engineering & Science
4	Dr. Jennifer So-kuen Chan	University of Sydney	School of Mathematics and Statistics

3 years (15 August 2017 – 14 August 2020)

Volatility is a measure of the instantaneous variability for financial assets. This project develops a new class of quantile range-based volatility measures as well as a new class of volatility models to account for the stylized facts that fit the volatility estimates to provide more accurate volatility forecasts using high frequency financial data. These fitted volatility estimates are then incorporated into return models to capture the heteroskedasticity of returns. Apart from that, a regime-switching return-based model is formulated to forecast the dynamics of the volatility and return. Different value-at-risk (VaR) and conditional VaR return forecasts based on these models are provided and tested.

1. Propose a new class of quantile range-based measures (estimators) for volatility.
2. Exploit the efficiency of the new range-based measures in (i) to forecast its volatility. Propose the enhanced range-based models incorporating different asymmetric mean function specifications and error distributions for investigating its forecasting precision.
3. Assess the performances of the proposed methodologies indicated in parts (i) and (iii) on the proposed enhanced volatility models.
4. Explore and extend the potential applications that the proposed volatility models are capable of in financial applications.

1. Talent:
   • 1 PHD
     1. Tan Shay Kee (Graduated)
   • 1 Master
     1. Tan Chia Yen (Graduated)
2. Publication:
   

   Target	Current
   4	4

   
   
   1. Article in Indexed Journals
      • Quantile range-based volatility measure for modeling and forecasting volatility using high frequency data (2019) - Web of Science (WoS)
      • On the speculative nature of cryptocurrencies: A study on Garman and Klass volatility measure (2020) - Web of Science (WoS)
      • Bayesian return forecasts using realized range and asymmetric CARR model with various distribution assumptions (2019) - Web of Science (WoS)
      • Dynamic volatility modeling of Bitcoin using time-varying transition probability Markov-switching GARCH model (2021) - Web of Science (WoS)

Statistical analysis of time series is an essential component of much of the economic and scientific endeavor of developed countries like Malaysia. However, Malaysia financial industry needs better and more accurate forecasting approaches in order to minimize associated risks. Hence, the methods developed in this project will tend to integrate the underlying techniques to cater for the fast changing in our economic environments. One of the aims in this project is to enhance and expand the knowledge in this domain of research through scholarly publications as well as to expose the post-graduate students to the relevant issues in practice. Consequently, with collaboration between the researchers both from Malaysia and Australia, this project which lies at the heart of the Strategic Research Priority intends to deliver the required skills and to sustain both countries economic competitiveness and advantages. The establishment of the approaches will eventually be beneficial, among others, to financial institutions and researchers, and will also draw out social implications by underpinning the long-term viability of both nations’ natural resources such as in servicing and manufacturing industries that could be related to environmental issues.