cuda-options-calc

Using CUDA to explore options pricing and modeling

Don't sue me

I am neither proficient with CUDA/C nor am I some sort of market wiz. In fact, this is my first time using CUDA. (I recommend it)

Early Days

This is the early days of this project. My goal is to resolve some unanswered questions I have with options and options pricing.

Which is better: Long or Short?

Over the long run, which is more lucrative? Is Theta/Vega Gang the way to be, or can Vega/Gamma pop the price of long options in some predictable fashion?

Given volatility and deviation of price over time, what is the overall probability of profiting from a trade?

I don't just mean "how likely is it that it lands between two prices?" I mean this:

1. Collect the probability of landing on every possible price (down to dollar or penny) - monte carlo method
2. Multiply that probability by the P/L of the option if it were to land at that price
3. Net total all possible P/L weighted by their probability

Is it more lucrative to manage options early, or let them expire?

I see no reason why I can't model the price of an option and the expected range of movement of the underlying given its volatility, and search through time to see if there is a particularly fruitful zone of time within that option's lifespan where I could plan to manage the winners or losers.

Will this research affect my performance, trading mechanics, psychology/confidence as I trade live?

We'll see!