Using VaR and Greeks to Assess Options Portfolio Risk
Summary
The document asks how to compare the risk and return of options portfolios, noting that standard performance measures, market-risk measures, and option sensitivities capture different aspects of risk. It highlights a difficulty with comparing options by delta alone: an out-of-the-money option may have lower delta than an in-the-money option while showing larger percentage price changes when the underlying moves. Time decay and portfolio margin requirements are also raised as considerations.
The response describes a division of practice: risk management teams at banks and hedge funds primarily use value at risk to summarize portfolio exposure, while desk traders inspect individual Greeks for a more detailed view. VaR is presented as indirectly capturing the combined effects of Greeks, whereas separate Greeks help show how exposures arise. The answer is brief and experiential; it gives no calculation method, validation evidence, treatment of margin, or discussion of VaR’s limitations for options portfolios, so it does not establish one measure as sufficient for choosing between portfolios.
Key ideas
- Option Greeks describe price sensitivity to underlying market factors, while VaR summarizes overall portfolio exposure.
- Risk teams may use VaR for an aggregate view and traders may monitor individual Greeks for detail.
- Delta alone may not capture relative return risk across options with different prices and moneyness.
- Theta and margin requirements can matter when evaluating options strategies.
- The response does not explain VaR calculations or establish that VaR alone is sufficient.
Tags
Full text
# What are the some good measures of risk for options? # What are the some good measures of risk for options? I've seen a number of measures of risk in my reading: Sharpe, Sortino, Calmar, etc. In CAPM there is Beta, and I've seen papers discussing how to modify CAPM for asymmetry. There is Value at Risk and discussion about its failings for distributions with fat tails, and how to compensate. Of course, for options, there are the Greeks - measures of how the price of an option changes w.r.t. factors - underlying (delta), volatility (vega), interest (rho) and of course there are second order Greeks to consider. But, for example "riskier" options - e.g. out-of-the-money (OTM) options that have higher chance of expiring worthless, have a lower delta, than "less risky" options - e.g. in-the-money (ITM) which are less likely to expire worthless. Does it make sense to consider $delta/ option$ $price$? For example, if the underlying moves, an ITM option will move more than an OTM option, but in terms of its return (the percentage or log change of the option), the OTM might be moving dramatically more. Then there is theta... just the risk over time of the option losing value. I've read some papers that talk about the 'returns' of various strategies, but few go into detail about the margin requirements. Specifically, recently, I've been looking at various option portfolios, but find it difficult to really get a good sense of how to measure their return vs risk. The return part is easy, the risk - not so much. So how do I choose which portfolio is "better"? Are there any good resources where I can find discussion of this issue? How do hedge funds, for example, typically measure their risk when it comes to portfolios including options? Apologies for the general question - is it too general/subjective? ## Answer by dm63 (score 5) https://quant.stackexchange.com/a/25507 I don't have a reference for you but I have some experience. Risk management departments at hedge funds and banks would primarily look at the Var in order to capture the risk of an options portfolio. The var indirectly captures all the Greeks in a single measurement , since each Greek generates some exposure. The desk traders would tend to look at all the Greeks individually rather than the var, since they need a more detailed picture.
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.