Forecasting realized volatility: HAR against Principal Components Combining, neural networks and GARCH
Dimitrios I. Vortelinos
Research in International Business and Finance, 2017, vol. 39, issue PB, 824-839
This paper examines whether nonlinear models, like Principal Components Combining, neural networks and GARCH are more accurate on realized volatility forecasting than the Heterogeneous Autoregressive (HAR) model. The answer is no. The realized volatility property of persistence is too important to leave out of a realized volatility forecasting model. However, the Principal Components Combining model is ranked very close to HAR. Analysis is implemented in seven US financial markets: spot equity, spot foreign exchange rates, exchange traded funds, equity index futures, US Treasury bonds futures, energy futures, and commodities options.
Keywords: HAR; Principal Components Combining; Neural networks; GARCH; Forecasting (search for similar items in EconPapers)
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
Full text for ScienceDirect subscribers only
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:eee:riibaf:v:39:y:2017:i:pb:p:824-839
Access Statistics for this article
Research in International Business and Finance is currently edited by T. Lagoarde Segot
More articles in Research in International Business and Finance from Elsevier
Series data maintained by Dana Niculescu ().