RISIKO PASAR: PERBANDINGAN MODEL EWMA DAN GARCH PADA NILAI TUKAR RUPIAH TERHADAP US DOLLAR

Abstract

Financial risk model evaluation or backtesting is a key part of the internal model’s approach to market risk management as laid out by the Basle Committee on Banking Supervision. Using daily exchange rate from January 2006-February 2008, will be compared measuring volatility between EWMA (Exponential Weighted Moving Average) and GARCH (Generalized Autoregressive Conditional Heterocedasticity). The results show that GARCH methods have considerably better power properties in measuring the volatility than the EWMA methods. However, the number of exceptions from the GARCH model, although much less than the EWMA model but the numbers were still above 5% and 1% (confidence level of 95% and 99%). The arguments for explained this finding is a pressure from stakeholders or the existence of an economic events that result in changes in exposure due to the different policies. As a result, the VaR model would be inaccurate to reality.Keywords: volatility, backtesting, EWMA, and GARCH