with all due respect, you may want to read up on basic option theory in case this is a project that warrants such time investment. If you assume margin as max risk of an option you will later be in huge trouble should you ever continue to be exposed to the field of options pricing and risk management.
Zzzz1, I'm happy to explain on the phone. I'm just trying to help him with his question without judging his reasons. reg-t is rules based vs PM which is risk based. Except for short calls eg, a 1x2 call spreads-long 1 and short 2, reg-t would typically calculate max loss.
Zzzz1: You fail to comprehend the question! I could attempt to explain the basis of the question, but your prior responses imply that effort would be in vain as well.
it is ok, I was trying to help but seems you got what you were looking for, whatever it may be... maybe next time you could try to elaborate or be a bit more precise with the English language because "Amend question to be "I need algorithm for quantifying max Risk for DEFINED RISK complex multilegged option position" Note: Infinite is considered undefined! Think of trades possible in an IRA account." at least to me makes it very hard to understand what you are actually after.
There isn't enough information to answer your question accurately. For example, if your options are path dependent, the only way to get a reasonable estimate of your risk is to do some sort of Monte Carlo analysis. You also need to take all the hedges into consideration. If you really have hedged, then you have things like IRs that you have hedged, and underlying that is also in the portfolio. Basically, you want some sort of VAR for options that incorporates Monte Carlo. What was stated above about using a program that knows how to pair spreads and compute margin is correct and very important, especially if you are a MMer. It is just not complete. Excel add-in "Solver" is very popular for non-programmers in this space and does a pretty good job. You have to know how to set up the equations.
I've done several ways: - Stepandfetchit's sweep, but only on the stock price and at current time: gives some worst and best case scenarios. - Monte Carlo simulation of stock paths, along with delta hedging: this is the most time consuming but also gives the most info. - Use the greeks an extrapolate using current stock, time, vol etc: the info is only useful close around the current point and gets very innacurate when close to expiry (within a few days).
I use it in my trade tracking currently. See RISK and ROR added to my trade tracking below (which is based on the new equations). probably too much information...