ABSTRACT

Recently, modeling and forecasting of high-frequency data (such as daily price) volatility using GARCH-MIDAS attract the attention of many researchers. Thus, the objective of this study is to model the average daily coffee price volatility from 1 January 2010 to 30 June 2019. The GARCH-MIDAS component model decomposes the conditional variance into short run component which follows a mean-reverting unit GARCH process and long-run component which consider different frequency macroeconomic indicators via mixed interval data sampling (MIDAS) specification. Unit root test results show the return series are stationary at level, while macroeconomic variables are stationary at first difference except interest rate, which is stationary at level. From the result of estimated model, all selected indicators are crucial in explaining price volatility. . Moreover, the estimated GARCH-MIDAS model with money supply as a main driver is used for out-sample forecast. Based on, DM test statistic multiplicative GARCH-MIDAS model provides an explanation for stylized facts that cannot be captured by standard GARCH model.

摘要

最近，使用GARCH-MIDAS对高频数据（例如每日价格）波动进行建模和预测吸引了许多研究人员的注意力。因此，本研究的目的是对2010年1月1日至2019年6月30日的平均每日咖啡价格波动进行建模。GARCH-MIDAS组件模型将条件方差分解为短期组件，该组件遵循均值回复单元GARCH过程和长期运行组件，通过混合间隔数据采样（MIDAS）规范考虑了不同的频率宏观经济指标。单位根检验结果表明，收益率序列在水平上是平稳的，而宏观经济变量在除利率水平不变的利率之外的第一个差异处是平稳的。根据估计模型的结果，所有选定的指标对于解释价格波动至关重要。。此外，以货币供应量为主要驱动力的估计GARCH-MIDAS模型用于样本外预测。在此基础上，DM测试统计量乘以GARCH-MIDAS模型为标准GARCH模型无法捕获的风格化事实提供了解释。