Journal Article
No. 2009-8 | March 25, 2009
Forecast Evaluation of Explanatory Models of Financial Variability


A practice that has become widespread and widely endorsed is that of evaluating forecasts of financial variability obtained from discrete time models by comparing them with high-frequency ex post estimates (e.g. realised volatility) based on continuous time theory. In explanatory financial variability modelling this raises several methodological and practical issues, which suggests an alternative approach is needed. The contribution of this study is twofold. First, the finite sample properties of operational and practical procedures for the forecast evaluation of explanatory discrete time models of financial variability are studied. Second, based on the simulation results a simple but general framework is proposed and illustrated. The illustration provides an example of where an explanatory model outperforms realised volatility ex post.

Data Set

  • data_economics_ejournal_2009_8.txt
    txt file containing weekly exchange rates (USD/EUR, YEN/EUR, GBP/EUR and NOK/EUR) [text/plain; charset=US-ASCII, 7K]
  • readme.txt
    readme for the data in txt file [text/plain; charset=US-ASCII, 2K]

JEL Classification:

C52, C53, F31, F37, F47


  • Downloads: 5085 (Discussion Paper: 3290)


Cite As

Genaro Sucarrat (2009). Forecast Evaluation of Explanatory Models of Financial Variability. Economics: The Open-Access, Open-Assessment E-Journal, 3 (2009-8): 1–33.

Comments and Questions