Forecasting value at risk and conditional value at risk using option market data

Annalisa Molino, Carlo Sala

Research output: Indexed journal article Articlepeer-review

3 Citations (Scopus)

Abstract

We forecast monthly value at risk (VaR) and conditional value at risk (CVaR) using option market data and four different econometric techniques. Independent from the econometric approach used, all models produce quick to estimate forward-looking risk measures that do not depend from the amount of historical data used and that, through the implied moments of options, better reflect the ever-changing market scenario. All proposed option-based approaches outperform or are equally good to different “traditional” forecasts that use historical returns as input. The extensive robustness of our results shows that the real driver of the better forecasts is the use of option market data as inputs for the analysis, more than the type of econometric approach implemented.

Original languageEnglish
Pages (from-to)1190-1213
Number of pages24
JournalJournal of Forecasting
Volume40
Issue number7
DOIs
Publication statusPublished - Nov 2021
Externally publishedYes

Keywords

  • backtests
  • conditional value at risk
  • option market data
  • value at risk

Fingerprint

Dive into the research topics of 'Forecasting value at risk and conditional value at risk using option market data'. Together they form a unique fingerprint.

Cite this