Mixtures of t-distributions for finance and forecasting

Raffaella Giacomini, Andreas Gottschling, Christian Haefke, Halbert White

Research output: Contribution to journalArticlepeer-review


We explore convenient analytic properties of distributions constructed as mixtures of scaled and shifted t-distributions. Particularly desirable for econometric applications are closed-form expressions for antiderivatives (e.g., the cumulative density function). We illustrate the usefulness of these distributions in two applications. In the first application, we produce density forecasts of U.S. inflation and show that these forecasts are more accurate, out-of-sample, than density forecasts obtained using normal or standard t-distributions. In the second application, we replicate the option-pricing exercise of Abadir and Rockinger [Density functionals, with an option-pricing application. Econometric Theory 19, 778-811.] and obtain comparably good results, while gaining analytical tractability.

Original languageEnglish (US)
Pages (from-to)175-192
Number of pages18
JournalJournal of Econometrics
Issue number1
StatePublished - May 2008


  • ARMA-GARCH models
  • Forecast accuracy
  • Neural networks
  • Nonparametric density estimation
  • Option pricing
  • Risk-neutral density

ASJC Scopus subject areas

  • Economics and Econometrics


Dive into the research topics of 'Mixtures of t-distributions for finance and forecasting'. Together they form a unique fingerprint.

Cite this