On the economic benefit of utility based estimation of a volatility model

Adam Clements and Annastiina Silvennoinen ()

No 44, NCER Working Paper Series from National Centre for Econometric Research

Abstract: Forecasts of asset return volatility are necessary for many financial applications, including portfolio allocation. Traditionally, the parameters of econometric models used to generate volatility forecasts are estimated in a statistical setting and subsequently used in an economic setting such as portfolio allocation. Differences in the criteria under which the model is estimated and applied may inhibit reduce the overall economic benefit of a model in the context of portfolio allocation. This paper investigates the economic benefit of direct utility based estimation of the parameters of a volatility model and allows for practical issues such as transactions costs to be incorporated within the estimation scheme. In doing so, we compare the benefits stemming from various estimators of historical volatility in the context of portfolio allocation. It is found that maximal utility based estimation, taking into account transactions costs, of a simple volatility model is preferred on the basis of greater realized utility. Estimation of models using historical daily returns is preferred over historical realized volatility.

Keywords: Volatility; utility; portfolio allocation; realized volatility; MIDAS (search for similar items in EconPapers)
JEL-codes: C10 C22 G11 G17 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-ecm, nep-pke and nep-upt
Date: 2009-07-21
View list of references

Downloads: (external link)
http://www.ncer.edu.au/papers/documents/WPNo44.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:qut:auncer:2009_57

Access Statistics for this paper

More papers in NCER Working Paper Series from National Centre for Econometric Research
Contact information at EDIRC.
Series data maintained by School of Economics and Finance ().