In probability theory and directional statistics, a wrapped probability distribution is a continuous probability distribution that describes data points that lie on a unit n-sphere. In one dimension, a wrapped distribution will consist of points on the unit circle. If φ is a random variate in the interval (-∞,∞;) with probability density function p(φ), then z = e i φ will be a circular variable distributed according to the wrapped distribution pzw(z) and θ=arg(z) will be an angular variable in the interval (-π,π ] distributed according to the wrapped distribution pw(θ).
- in some interval of length
which is a periodic sum of period . The preferred interval is generally for which
In most situations, a process involving circular statistics produces angles () which lie in the interval from negative infinity to positive infinity, and are described by an "unwrapped" probability density function . However, a measurement will yield a "measured" angle which lies in some interval of length (for example ). In other words, a measurement cannot tell if the "true" angle has been measured or whether a "wrapped" angle has been measured where a is some unknown integer. That is:
If we wish to calculate the expected value of some function of the measured angle it will be:
We can express the integral as a sum of integrals over periods of (e.g. 0 to ):
Changing the variable of integration to and exchanging the order of integration and summation, we have
where is the pdf of the "wrapped" distribution and a' is another unknown integer (a'=a+k). It can be seen that the unknown integer a' introduces an ambiguity into the expectation value of . A particular instance of this problem is encountered when attempting to take the mean of a set of measured angles. If, instead of the measured angles, we introduce the parameter it is seen that z has an unambiguous relationship to the "true" angle since:
Calculating the expectation value of a function of z will yield unambiguous answers:
and it is for this reason that the z parameter is the preferred statistical variable to use in circular statistical analysis rather than the measured angles . This suggests, and it is shown below, that the wrapped distribution function may itself be expressed as a function of z so that:
where is defined such that . This concept can be extended to the multivariate context by an extension of the simple sum to a number of sums that cover all dimensions in the feature space:
where is the th Euclidean basis vector.
Expression in terms of characteristic functions
Using the delta function, a general wrapped distribution can be written
Exchanging the order of summation and integration, any wrapped distribution can be written as the convolution of the "unwrapped" distribution and a Dirac comb:
The Dirac comb may also be expressed as a sum of exponentials, so we may write:
again exchanging the order of summation and integration,
By analogy with linear distributions, the are referred to as the characteristic function of the wrapped distribution (or perhaps more accurately, the characteristic sequence). This is an instance of the Poisson summation formula and it can be seen that the Fourier coefficients of the Fourier series for the wrapped distribution are just the Fourier coefficients of the Fourier transform of the unwrapped distribution at integer values.
The moments of the wrapped distribution are defined as:
Expressing in terms of the characteristic function and exchanging the order of integration and summation yields:
From the theory of residues we have
where is the Kronecker delta function. It follows that the moments are simply equal to the characteristic function of the unwrapped distribution for integer arguments:
Generation of random variates
If X is a random variate drawn from a linear probability distribution P, then will be a circular variate distributed according to the wrapped P distribution, and will be the angular variate distributed according to the wrapped P distribution, with .
where is any interval of length . If both the probability density and its logarithm can be expressed as a Fourier series (or more generally, any integral transform on the circle) then the orthogonality property may be used to obtain a series representation for the entropy which may reduce to a closed form.
The moments of the distribution are the Fourier coefficients for the Fourier series expansion of the probability density:
If the logarithm of the probability density can also be expressed as a Fourier series:
Then, exchanging the order of integration and summation, the entropy may be written as:
Using the orthogonality of the Fourier basis, the integral may be reduced to:
For the particular case when the probability density is symmetric about the mean, and the logarithm may be written:
and, since normalization requires that , the entropy may be written:
||This article includes a list of references, but its sources remain unclear because it has insufficient inline citations. (July 2011) (Learn how and when to remove this template message)|
- Mardia, Kantilal; Jupp, Peter E. (1999). Directional Statistics. Wiley. ISBN 978-0-471-95333-3. Retrieved 2011-07-19.
- Mardia, K. (1972). Statistics of Directional Data. New York: Academic press.
- Borradaile, Graham (2003). Statistics of Earth Science Data. Springer. ISBN 978-3-540-43603-4. Retrieved 31 December 2009.
- Fisher, N. I. (1996). Statistical Analysis of Circular Data. Cambridge University Press. ISBN 978-0-521-56890-6. Retrieved 2010-02-09.
- Circular Values Math and Statistics with C++11, A C++11 infrastructure for circular values (angles, time-of-day, etc.) mathematics and statistics