Selecting Option Trades by Improving Portfolio Greeks
Summary
The document outlines a portfolio-based approach to deciding which proposed option trades to accept. For options sharing an underlying, it recommends calculating aggregate sensitivities such as delta, gamma, and vega, then comparing them with target values or acceptable ranges. A candidate trade is favored when its Greeks move the portfolio toward those targets; for example, if portfolio delta is already above its desired level, a trade with negative delta may help while a positive-delta trade may worsen the imbalance.
The answer notes that a portfolio spanning several underlyings requires tracking sensitivities for each one, making the problem more complex. The proposal is a high-level screening principle rather than a fully specified optimization algorithm: it does not describe trade prices, risk limits beyond the mentioned Greek targets, competing objectives, or how to rank trades when several are available. The document offers conceptual guidance, not performance evidence or a tested system.
Key ideas
- Calculate portfolio Greeks such as delta, gamma, and vega before assessing proposed option trades.
- Define target values or acceptable ranges for the portfolio’s risk sensitivities.
- Prefer trades that move current exposures toward the chosen targets.
- Track separate exposures when options reference different underlying assets.
- The suggested rule is conceptual and does not specify pricing, optimization, or trade-ranking details.
Tags
Full text
# Portfolio Analysis Interview Question # Portfolio Analysis Interview Question Suppose you have a portfolio of 100 options. Then I give you a subset of trades in which you can make. The trades consist of possible buys/sells of different options from different clients. Discuss how would you design a trading system/algorithm to determine which trades to accept and which to deny? Hint: Use portfolio theory. ## Answer by Alex C (score 4) https://quant.stackexchange.com/a/41354 The question is pretty vague. Don't let that get to to you and just take a stab at some sort of answer: Assuming all the options are on the same underlying (say the S&P 500), the option portfolio will have overall stats such as Delta, Gamma, Vega (perhaps a few others) which can be easily computed. The managers will also have targets in mind for these stats (either as optimal values $\delta^*,\gamma^*,\nu^*$ or perhaps as ranges which they consider acceptable e.g. $\delta_L\le \delta\le \delta_H$ etc.). The algorithm to review possible trades would work like this: You accept a trade if it brings the parameters of the portfolio closer to the desired values. For example if Delta of the portfolio is already too high, then you do not accept any trades with a positive delta, but you do accept negative delta trades, which help to bring the portfolio delta down. Some details would have to be filled in for the algorithm to be exactly specified, but this would be the basic idea. If the options are on different underlyings it gets more complicated, we might have to keep track of multiple Deltas etc. with respect to multiple underlyings. ## Answer by vanguard2k (score 2) https://quant.stackexchange.com/a/41381 The question is vague on purpose. I think that the interviewer just wanted to hear what you know about options and how they work in combination with each other. This way, he will get a feeling about how well you can handle these securities and portfolios. The amount of things you could say about this type of scenario is sheer endles...
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.