Stock Price Probability with Stable Distributions

Requires a Wolfram Notebook System

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

Requires a Wolfram Notebook System

Edit on desktop, mobile and cloud with any Wolfram Language product.

This Demonstration calculates the probability that the random price of the exchange-traded fund, SPY, will be higher or lower after one month than a particular future possibility.


Stable distributions are controlled by four parameters, , , , . When the shape parameter , the distribution is normal; when , the distribution has infinite variance; and when , the distribution has no mean. The other parameters are , the skewness parameter; , the scale parameter; and , the location parameter, which is the mean of the distribution when .

The calculation uses a stable distribution model, with parameters , , , derived from intraday data from the last mid 2007 to December 2008.

The calculation is based on the last price, which you must set first either with the slider or by directly entering the price. You can then move the future price slider to calculate the probabilities along the curve for various future prices.

Warning! This is an example only; please do not make stock purchases based on these calculations.


Contributed by: Bob Rimmer (March 2011)
Open content licensed under CC BY-NC-SA



Levy-stable distributions were introduced for financial returns by Mandelbrot in 1963. They offer an advantage over the log normal assumption in that extreme events can exist in the model. They have not been in much use because of the difficulty in performing the calculations. Most models using stable distributions have also assumed that they are stationary; when parameters are fitted to a stationary stable model, the tails are consistently found to be too heavy. The parameters for the particular example were scaled from intraday observations at one-minute intervals. The parameters chosen for the example were obtained by maximum-likelihood fitting (using an FFT approximation of the stable density from the sampled stable characteristic function) over six months of data, after rescaling each day's data by a value of derived from an empirical characteristic fit. This method removes most of the serial dependence seen in the data and gives a value of that is significantly higher than would be found by fitting the raw data, assuming the distribution is stationary. The parameters and are selected from average days, but the user needs to be aware that may vary by a factor of 2 or 3 during volatile periods in the market. The user who downloads the source may experiment with his own parameters.

Since this graphic was first put up the stock market has demonstrated some remarkable behavior. Below is a plot of intra-day gamma calculations from July 2007 to mid December 2008, rescaled to one month to help in guessing the gamma parameter for the future. This new version of this Demonstration allows the user to adjust the gamma parameter as a measure of volatility in the market.

Software is now available for Mathematica that can do the parameter fitting and other calculations. This Demonstration uses only a fragment of the free software available at mathestate, where there is a growing series of web pages devoted to the financial applications of stable distributions, including a nonstationary stable model.

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.