Hello!
I am working on comparing valuations of vanilla call options using either LIBOR with historical dividends or OIS with implied dividends in the Black Scholes formula. Unlike historical dividends, the implied dividend rate shows more volatility. I was wondering if the forum had any insight into smoothing techniques that could be applied to this figure to achieve a more stable valuation over time.
Many thanks!
Serving the Quantitative Finance Community
Re: Smoothing Implied Dividends in Black Scholes
There are single-name equities and broad-based indices.
1. Single-name equities.
Here the dollar dividends are discrete. I wouldn't use a 'rate'. Treating discrete dollar dividends sensibly under Black-Scholes is both tricky and contentious: there are various approaches. The one I advocate is given in Ch. 9 ('Back to Basics: an update on the discrete dividend problem') in my book "Option Valuation under Stochastic Volatility II". Before reading that chapter, you might also want to take a look at the 2003 article from Wilmott magazine, similarly titled "Back to basics: ..." by Haug, Haug, and Lewis. (available in various places).
It's possible your implied dividends are not behaving sensibly because you are using an older (improper, IMO) escrowed dividend model. The problems with escrowed dividends are discussed in both of my sources for you.
2. Broad-based indices.
Here, a time-varying dividend rate makes sense to extract. To do so, I recommend extracting an inferred dividend rate by the method of the "VIX white paper". The method amounts to fixing an interest rate and then extracting an option-implied forward price for the underlying, with the maturity of the forward contract identical to the maturity of the options. Of course, if it is for an important purpose, you can always try to reconcile that rate with direct day-by-day dividend projections.
1. Single-name equities.
Here the dollar dividends are discrete. I wouldn't use a 'rate'. Treating discrete dollar dividends sensibly under Black-Scholes is both tricky and contentious: there are various approaches. The one I advocate is given in Ch. 9 ('Back to Basics: an update on the discrete dividend problem') in my book "Option Valuation under Stochastic Volatility II". Before reading that chapter, you might also want to take a look at the 2003 article from Wilmott magazine, similarly titled "Back to basics: ..." by Haug, Haug, and Lewis. (available in various places).
It's possible your implied dividends are not behaving sensibly because you are using an older (improper, IMO) escrowed dividend model. The problems with escrowed dividends are discussed in both of my sources for you.
2. Broad-based indices.
Here, a time-varying dividend rate makes sense to extract. To do so, I recommend extracting an inferred dividend rate by the method of the "VIX white paper". The method amounts to fixing an interest rate and then extracting an option-implied forward price for the underlying, with the maturity of the forward contract identical to the maturity of the options. Of course, if it is for an important purpose, you can always try to reconcile that rate with direct day-by-day dividend projections.
Re: Smoothing Implied Dividends in Black Scholes
Thank you for the reply! I am attempting to value call options on a broad based index, specifically the S&P500. Another method that was referred to me took the dividend future index value less a repo rate but I am not exactly sure where to source the appropriate repo rate for this method. Has anyone utilized this methodology before?
Re: Smoothing Implied Dividends in Black Scholes
I don't know, but I see there is an associated SPXDIV index, which should help you to make very accurate dividend projections, assuming the horizon is not too distant.
-
- juliandenton
- Posts: 1
- Joined:
Re: Smoothing Implied Dividends in Black Scholes
I hope that someone was able to leave the most appropriate response. As far as I could see, there are no options for all of us to choose.