How do I transform a discrete sample into a continuous distribution.
Let's suppose I have a collection of points (time, price of stock) for instance. I would like to transform it into a distribution (price of stock, probability). How do I do that? Should I need to use the Central Limit Theorem, even if the distribution probably won't be a normal? I mean, I would not like to normalize it.
For small amount of data, it's fairly easy to do it by hand and then graph. However, for a gigantic sample, it's almost impossible. Thanks