In recent years, financial markets have faced with high volatility so that the unreliability caused by this, has made investors to be concerned. Therefore, volatility modeling and prediction is taken into consideration in various researches and practical issues. I
More
In recent years, financial markets have faced with high volatility so that the unreliability caused by this, has made investors to be concerned. Therefore, volatility modeling and prediction is taken into consideration in various researches and practical issues. In this regard, access to high frequency data creates a new field in volatility modeling and return prediction of financial assets.
In this Thesis using high frequency data, the volatility modeling has been performed by HAR-RV family models, and the effect of adding jump component to volatility prediction efficiency of index is studied in Tehran Securities Exchange. The results of modeling and prediction suggest that the forecast error is decreased by adding jump component into model, and also the separation of jump and continuous components of realized volatility is effective on performance improvement.
Manuscript profile
