2021, Vol. 2, Issue 1, Part A
A study of conditional volatility of hybrid FIGARCH, and Midas regression
Author(s): MU Bawa, Dr. HG Dikko, Dr. J Garba and Dr. S Sadiku
Abstract: This study is to discuss the techniques that will be employed by the researcher’s when conducting the study on modelling and predicting financial Time Series data. The hybridization between Fractionally Integrated Generalized Autoregressive Conditional Heteroscedastic (FIGARCH) processes. With MIDAS Regressions will be used to develop the most appropriate model for forecasting financial Time Series data. However, same as the main weakness of the original FIGARCH model, it assumes that the conditional volatility has only one regime over the entire period. Unfortunately, it is not always true. Which can lead to spurious regression. Many studies have suggested that structural breaks should be combined into the long memory models to properly ﬁt ﬁnancial return volatility (Baillie and Morana, 2009; Belkhouja and Boutahary, 2011). In order to overcome such draw back the distant feature of the new class is that mixed data sampling allow us to link volatility directly with economic activity (i.e. data that is typically sampled at the different frequency than daily returns. With the component of long-run ( ) of the mixed data sampling (MIDAS) can change over a specific time that is daily or monthly and involves rolling windows of Financial data.
Pages: 13-20 | Views: 623 | Downloads: 335
Download Full Article: Click Here
How to cite this article:
MU Bawa, Dr. HG Dikko, Dr. J Garba and Dr. S Sadiku. A study of conditional volatility of hybrid FIGARCH, and Midas regression. Journal of Mathematical Problems, Equations and Statistics. 2021; 2(1): 13-20.