Handling Discrete Dividends and Limitations in OptionMetrics Data

Summary

The document raises practical concerns about using OptionMetrics data to compare option pricing models, especially whether European call valuations should account for discrete dividends. It asks whether the vendor makes this adjustment and what other data corrections may be needed, but it does not provide a detailed workflow or documentation guide.

The response recommends treating OptionMetrics primarily as a source of option prices, then estimating implied volatilities with a chosen model and independently sourced dividend inputs. It also cautions that dividend data may be unreliable, subtracting discounted dividends can misrepresent cash flows, borrow and lending spreads are not controlled for, and the vendor's American option model may be inadequate for research. These are the respondent's assessments rather than a comparative study: no supporting tests, alternative model specifications, or validation results are presented. Researchers should therefore treat the advice as a warning to inspect assumptions and validate inputs, not as evidence that every OptionMetrics field is unsuitable.

Key ideas

• Discrete dividends can affect option valuation and may require explicit treatment when comparing pricing models.
• The response questions the quality of the vendor's dividend information.
• Subtracting discounted dividends is described as an imperfect way to represent cash flows.
• Borrow and lending spreads may be absent from the vendor's pricing assumptions.
• Researchers can estimate implied volatility from option prices using independently selected models and dividend inputs.

Tags

# Question on OptionMetrics: when are adjustments for discrete dividends needed?


# Question on OptionMetrics: when are adjustments for discrete dividends needed?












Bakshi et. al. (1997) analyzes the empirical performance of some alternative option pricing models. I am interested to do this as well - hence applying different models - but I am unsure how to handle the OptionMetrics data.

Something that stood out from the article for instance is that they adjust the spot stock price for discrete dividends of European call options. My first question is then immediately, when does this need to be done? And does anybody knows whether OptionMetrics does this automatically. Also which other adjustments (beside the filters) are needed? How can I know which adjustments need to be done and is there any documentation on this?

Additionally, does anybody know a good guide on starting to work with OptionMetrics (i.e. what to watch out for and how to handle the data). Thank you.

## Answer by Brian B (score 4)

https://quant.stackexchange.com/a/4587

I've always been skeptical of the quality of OptionMetrics' dividend information in the first place. Furthermore, subtracting out discounted dividends is an inaccurate way to deal with cashflows while pricing options.

OptionMetrics also does not control for borrow/lend spreads, and their American option pricing model is not the best.

My recommendation would be to use OptionMetrics only for price information, and back out your own volatilities using better-designed models and some good dividend information. OptionMetrics vols are good enough for the back office, but not really for serious work.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.