Cauchy distribution

Understanding the Cauchy Distribution

The Cauchy distribution, also known as the Lorentz distribution, is a continuous probability distribution that arises in various statistical contexts. It is named after the French mathematician Augustin-Louis Cauchy. Unlike the well-known Gaussian distribution, which is often used in statistics due to its nice mathematical properties, the Cauchy distribution is notorious for its "pathological" behaviors.

Definition of the Cauchy Distribution

The Cauchy distribution is defined by its probability density function (PDF), which has the following form:

f(x; x0, γ) = 1 / [πγ(1 + ((x - x0) / γ)^2)]

Here, x0 is the location parameter, which specifies the peak of the distribution, and γ (gamma) is the scale parameter, which describes the width of the distribution. The scale parameter is often referred to as the half-width at half-maximum (HWHM) because it is the half of the width of the peak at its half-maximum.

Properties of the Cauchy Distribution

One of the most striking properties of the Cauchy distribution is that it does not have a mean, variance, or higher moments defined. This is because its tails are much heavier than those of the Gaussian distribution, leading to integrals that do not converge when trying to calculate these statistics. As a result, the central limit theorem does not apply to the Cauchy distribution, and sums of independent Cauchy-distributed random variables have the same distribution as the individual random variables.

Despite its lack of a mean and variance, the Cauchy distribution does have a median and mode, both of which are equal to the location parameter x0.

Applications of the Cauchy Distribution

The Cauchy distribution appears in various areas of physics and engineering, particularly in scenarios involving resonance behavior. For instance, it describes the distribution of energy and momentum in systems undergoing resonant scattering. It also appears in the context of signal processing, where it can model the distribution of noise or the shape of spectral lines.

In finance, the Cauchy distribution has been used to model the distribution of returns on investment, as it can capture the heavy tails often observed in financial data better than the Gaussian distribution.

Challenges in Working with the Cauchy Distribution

Due to its heavy tails, the Cauchy distribution can be challenging to work with in statistical contexts. Estimating parameters of the Cauchy distribution using standard techniques like the method of moments is not possible, as these moments do not exist. Instead, parameter estimation is typically done using methods such as maximum likelihood estimation or Bayesian inference.

Moreover, because of its undefined mean and variance, many standard statistical tests and confidence intervals that rely on these moments cannot be used with Cauchy-distributed data. Specialized techniques and robust statistical methods are often required when working with such data.


The Cauchy distribution is a fascinating probability distribution with unique properties that distinguish it from other distributions. While it poses certain challenges in statistical analysis, it remains an important distribution due to its relevance in various scientific and engineering fields. Understanding and working with the Cauchy distribution requires a solid grasp of its characteristics and appropriate statistical methods tailored to its peculiarities.


For further reading on the Cauchy distribution and its applications, consider exploring the following resources:

  • Statistical textbooks that cover non-standard distributions and robust statistics.
  • Research papers in physics and finance that model phenomena with heavy-tailed distributions.
  • Articles on the history and contributions of Augustin-Louis Cauchy to mathematics and probability theory.

Please sign up or login with your details

Forgot password? Click here to reset