Just exactly what you are looking for... about $90 US. It uses a Monte Carlo generator. It will run simulations (I use one million sample paths) and tell you the likelyhood of exceeding lower or upper boundary given a certain volatility over a certain period of time. I hope you find it as valuable as I have. http://www.optionstrategist.com/products/analysis/software/index.html Peace and gtty, Lar
If you use IV and compute expected value you should get the fair option price value, that's how these models work. You only have an edge if real probabilities differ from the probabilities implied by the option price (or if you are an MM profiting from spreads). I use a conservative HV value and consider support/resistance levels to increase the probabilities.
Alassio, I know how the models work. The problem, though, is that you don't know the REAL probability. Besides, I don't trade to exploit "mis-pricings" between IV and HV. I only use a probability calculator to give me a rough idea of what the stock is capable of doing in a certain period of time. In any case, volatility is a big fudge-factor in options so I try to keep it as simple as posssible.
The problem with individual stocks is that they have much more fat tailed probability distributions than e.g. an index. The standard model can be quite misleading (e.g. 3 sigma moves happen quite often). If you want to use the probability distribution the market is expecting, you should use the probability distribution that is implied by the volatility skew. However, I haven't found yet a probability calculator that gives me that result. I would be interested, if anybody provides such a sophisticated probability calculator. My current approximation is a conservative approach (lacking a bether model): - if I want the stock to move, use the min of IV and HV - if I want it to sit still, use HV + Offset
I'm no statistics expert here... But be careful. If your rely on probability correlation alone... make sure your distribution is linear. If it is not, the correlation coefficient will understate the estimate, and in some cases, give a totally erroneous answer. Here this might help if your not familiar:
gbos I've just started to work those formulas you posted. I can't get any to work. I think the problem may be in the Snorm function ?
No, the formulas are working fine both in excel and in any basic compiler. In the zip file I have the implementation exe compiled with FreeBasic compiler together with source code. What error messages did you get?