Universitas Indonesia Conferences, International Accounting Conference - 2017

Font Size: 
Extreme Value Theory Value at Risk in High Frequency Data: Evidence from Australian Interconnected Power Markets
Rangga Handika, Rahfiani Khairurizka

Last modified: 2017-07-12

Abstract


This paper proposes extreme value theory (EVT) Value-at-Risk (VaR) in the Australian interconnected power markets. We develop one-day ahead EVT VaR estimates and perform the backtesting to demonstrate how the EVT VaR improves the performance of standard normal VaR. We document that EVT VaR is superior than normal VaR for all regions and at three different confidence levels. Therefore, we conclude that EVT VaR is more appropriate than the standard normal VaR in the Australian interconnected power markets.


Keywords


Power Markets, Extreme Value Theory, Value-at-Risk, Backtesting

Conference registration is required in order to view papers.