A while back, panzerman posted the following as a way to estimate future spot price as a function of volatility, deviation, and time: X = exp(sigma*t*x)*S where X = future spot sigma = percent volatility t = sqrt(days 'til expiry/365) x = standard deviations S = current spot What is the proper IV to use in the equation? I know it's annualized, as the root time term takes care of shorter expiry. But which annualized IV? ATM? -- puts, calls, average of the two? Some blended IV for the entire option chain? My broker reports a blended IV that is close to that average (i.e., on SPY it's currently reporting ~ 13.5, and the ATM calls show IV ~ 9.5 and ATM puts are at IV ~ 15-18 or so). Any help or direction on the derivation would be great. Thanks!
jimmyjazz ........ I hate to be the bearer of bad news, but someone has to step up to the plate otherwise your thread will end up with 0 replies. So here it goes ........ ...... Estimating the future price of a stock is nothing more than fantasy. Sure you can try and use some sort of formula but the end results will be irrelevant garbage that will require additional analysis - perhaps even with another formula. Put the option greeks on the back burner and get to the basics - study the charts and go with your gut feeling. Hope that helps.
No, it doesn't help. Please stop responding to my threads with your crap. It's an equation, and I'm simply asking about one variable. It's a statistical estimate which has some value, regardless of what you think. If it helps you sleep at night, imagine I'm trying to estimate the growth of honey bees in your boxers when you stuff your pants with clover.
I’m not sure about what panzerman wrote, but are you expecting an actual value or a range? Are you trying to calculate the forward price based on the listed options (which have vols)? And even if you calculate this “future spot price”, what are you hoping to achieve with it?
Actually, OTM is really helpful that he will ruffle the feathers or our sharpest option theorists and make them post. For now, we are all you have. And what you got here, sir, is a roscoe that won't be able to hit the side of a barn. Use the average of ATM put and call IV. Assume put-call parity. No dividends. European exp. Keep the model simple, so you can build concepts and intuitive understanding of it more easily.
It's a method for estimating spot price "x" standard deviations from current spot at some future time. Option pricing allows us to infer IV, and IV allows us to calculate the standard deviation. Go 1 StDev down from spot and you get an estimate of the price which should not be breached ~ 84% of the time, given whatever current and expected information got priced into IV. Go 1 StDev up, same thing, but in the other direction. Now we have an estimate of future range. Nobody is saying it's a rock solid prediction. It is an attempt to bound future behavior based on collective market information. (I'm sure I am mis-characterizing the actual implications, and I welcome any corrections that fall in line with the spirit of the thread. I don't need a lecture on the uselessness of Black-Scholes, Greeks, etc.)
A faster, more heuristic method for this is to simply say: +/- 1 Sigma ~ Spot +/- ATM Straddle. Problems arise though... 1) Using this method is equivalent to inferring a quick and dirty estimate for a +/- 1 SD range via ATM IV. However, in reality you're forgoing the use of a lot of other readily available data -- i.e. the other options! A far better method might be to use the method they use to calculate the VIX and apply it to your option series. This method calculates an IV number that incorporates the overall implicit risk-neutral density that is priced into the options. It will likely deviate from the value you get just using the ATM. 2) Per 1), in order to calculate these estimates, you're using IVs. IVs are derived from option pricing models. Because of this, they come from the risk-neutral world. You are trying to calculate future spot -- a real world variable. Not quite apples-to-apples. Whether or not this matters is debatable, but it's certainly worth being aware of.
Thanks, longthewings. I just scanned a CBOE white paper on calculating VIX. Nothing too crazy in terms of the math, but plenty of room for making mistakes. I might try to build that model in Excel and see if I can get it to match the VIX. At that point, I might have a solid model for predicting the right IV for the model I listed above. Until then, I'll just blend the ATM call and put IV. Thanks!
Last time I looked at the predictive power of the ATM straddle on SPY, it was in 2014 and about 67% of the time, the spot at expiration was within the range “predicted” by the ATM straddle 10 business days prior. Obviously, when spot expires outside...