Title: Option implied tail risk and expected stock returns
Authors: Conall OSullivan - University College Dublin (Ireland) [presenting]
Yan Wang - University College Dublin (Ireland)
Abstract: The use of a Cornish-Fisher expansion is proposed in order to estimate the value-at-risk and expected shortfall of a risk-neutral distribution using model-free option-implied moments. We extract these risk-neutral tail risk measures from SP-500 index options on a daily basis and average over each month to obtain a monthly time series of forward-looking value-at-risk and expected shortfall across different tenors. These risk-neutral tail measures predict the equity risk premium at longer horizons (6 months or more) and are robust to the inclusion of other option implied predictors such as the variance risk premium. The risk-neutral tail measures are a priced factor in the cross-section of stock returns as stocks with higher loadings on the RN tail measures earn lower returns on average in subsequent months. This is consistent with these stocks acting as a hedge against perceived tail risk.