Modeling Return Distributions

Initializing live version
Download to Desktop

Requires a Wolfram Notebook System

Interact on desktop, mobile and cloud with the free Wolfram Player or other Wolfram Language products.

We use numerical maximum likelihood estimation to obtain the parameters governing probability distributions that might plausibly generate the daily, weekly, monthly, or quarterly return distributions of stocks from the Dow 30. The candidate distribution types should probably be continuous and supported on the whole real line. So, we choose to investigate the Gaussian distribution, Student's -distribution, the Gumbel distribution, the Cauchy distribution, and the Laplace distribution. We show a histogram of the return data for a particular time horizon and frequency, as well as the density function with the fitted parameters.


Explore the various possible distribution choices to see which one(s) fit the empirical returns. Some choices for some return frequencies are clearly better than other choices!


Contributed by: Jeff Hamrick (March 2011)
Additional contributions by: Jason Cawley
Open content licensed under CC BY-NC-SA



In this Demonstration, we define the returns of a stock by for . Hence, the number corresponds to a positive 5% return and corresponds to a negative 9% return.

Feedback (field required)
Email (field required) Name
Occupation Organization
Note: Your message & contact information may be shared with the author of any specific Demonstration for which you give feedback.