Volatility Modelling In Crude Oil and Natural Gas Prices

摘要

This study analysis the return volatility of spot market prices of crude oil (WTI) and natural gas (Henry Hub) for two different terms which cover 02.01.2009-28.04.2014 and 04.01.2010-28.04.2014 with different version of the GARCH class models such as GARCH, IGARCH, GJRGARCH, EGARCH, FIGARCH, FIAPARCH. In particular, the main idea of employing various GARCH models is to determine which one of these linear and nonlinear asymmetric models perform more accurate in terms of ingroups and intergroups activities. Therefore, the main purpose of the paper is to determine a model which ensures to get a maximum return with response to the minimum loss for returns of the investments held by individual investors and fund managers, private sector budget planning decision makers, and state agencies forecasting about macroeconomic indicators. To do this, the ten-days out-of-sample volatility forecasts of Loss Functions to capture the forecasting performance of GARCH class models and to prevent forecasting errors with efficiency hedge ratio in energy market are being considered. For two periods, asymmetric and integrated GARCH models give relatively more accurate performance than other available models. Respectively, for the first period, minimum loss model is FIGARCH-BBM (SST) and for the second period, is EGARCH(GED) for WTI crude oil series in consideration of MSE and MAE criterion. Similarly, for the first period minimum loss model is FIGARCH-BBM (SST) and for the second period, is EGARCH(GED) for Henry Hub natural gas series in consideration of MSE and MAE criterion. This study has potential recommendations for investors from developed and developing countries, which differs it from the current studies.