Implied Distributions from GBPUSD Risk-Reversals and Implication for Brexit Scenarios

Implied Distributions from Risk-Reversals and Brexit/Trump Predictions

“…As already presented elsewhere in the literature [1], Figures 1 and 2 show implied densities for GBPUSD and EURGBP respectively from 1 May 2015 to 23 June 2016 for options expiring the day after the Brexit referendum date, and also the implied density as seen on Friday 24 June 2016 for the FX rate one business day later. These results suggest probability mass for GBPUSD between 1.10 and 1.30, and for EURGBP between 0.825 and 0.95 at the referendum date.…”

“…In the case of the French elections, no particular event risk analogous to Brexit came to pass, whereas the surprise factor of the Conservative Party's poor result in the June 2017 UK general election is consistent with our expectations based on implied distributions. It would be of interest to extend and strengthen this analysis by utilizing more quantitative methods such as the mixture model approach [1] and including additional predictive information such as betting market odds and polls data [10], which would allow us to strengthen the analysis, both in terms of describing market regimes and apportioning them to potential Innovations in Insurance, Risk-and Asset Management Downloaded from www.worldscientific.com by 44.224.250.200 on 08/03/20. Re-use and distribution is strictly not permitted, except for Open Access articles.…”

“…Note that the risk-free rate r d is determined separately for each maturity given the underlying yield curve, rather than being presumed constant. As well as the empirical approach based around the Breeden-Litzenberger analysis above, we construct a mixture model, as described in [1] and references therein, which has a specific probability P E of a risk event (such as a vote for Brexit, Trump, Le Pen or a vote against a Theresa May majority government) on the referendum/election date which we expect the market could well be sensitive to, and thereby associated with an event risk potentially likely to cause a jump in FX rates. We let T * denote the time to that referendum/election date and use P N E to denote the probability of no such risk event happening (clearly P E + P N E = 1).…”

“…As well as our own work [1], Dupire [9] and Hanke et al [10] use similar methods to model potential FX moves around the dates of the Brexit referendum and the 2016 US election (specifically considering the FX example of the Mexican peso quoted against the US dollar). Dupire [9] also discusses the 2017 French elections, with respect to equity markets, which have a traded market in volatility through the VIX [11].…”

“…Our contribution to this volume seeks to empirically investigate the implied probability distributions around these event dates. We have already considered the case of Brexit [1] and we shall review our analysis herein, and then extend the analysis to Trump and the 2017 elections mentioned above.…”