Jump to content

Draft:q-distribution

From Wikipedia, the free encyclopedia
  • Comment: This has reached the end of the list of drafts for review, which must surely mean that the article is too complex in the current form. See Gaussian function where at least the Lead section gives some background as to the use and significance of that function (though it is far from ideal), here there are just a few words. Not everyone will be able to walk into the formulae, but at least an educated reader should be able to get through the first few paragraphs. More supportive text would make this a much better article. For example amplifying the potential use in financial analysis, why and how this distribution could be used, examples of where it is not so useful.
    The wiki links in paragraph 2 go to disambiguation links, which in turn adds to the complexity. E.g. Exponential.
    The article will doubtless be improved by other editors after some fairly modest improvements, so I hope the submitting editor persists a little longer. ChrysGalley (talk) 12:32, 29 December 2025 (UTC)
  • Comment: New references appear to be searches for other distributions? This needs more sourcing for this in particular. Smallangryplanet (talk) 15:03, 11 December 2025 (UTC)
  • Comment: Only one source. Also it is overly complex. General readers won't have a clue what is going on here. Ktkvtsh (talk) 02:34, 6 December 2025 (UTC)
  • Comment: This article in its present form does not lack sufficient context, since it begins by saying "In the mathematical theory of probability" and its provision of context is comparable to that of Wikipedia's many other articles on particular probability distributions. Several problems pointed out by ChrysGalley are trivially easy to correct and I have done so: In particular, the link to Gaussian function was obviously the wrong target and I have fixed it. I also fixed the several links whose appropriate targets were other articles about particular probability distributions. The comment by Ktkvtsh is silly: an article about a family of probability distributions needs to be written in a way appropriate to the subject, it in this respect this article is comparable to Wikipedia's many articles about particular families of probability distributions. One thing the article conspicuously lacks is the usual probability distributions template. My remaining concern is that the only source is the paper in which this distribution was first introduced. Citing secondary sources establishes "notability", and that is not here. Michael Hardy (talk) 23:12, 1 January 2026 (UTC)

In the mathematical theory of probability, Zhao's q-distribution is any of the heavy-tailed probability distributions of a real-valued random variable belonging to a certain four-parameter family of distributions.[1] It is proposed for modeling financial asset returns, which usually have heavier tails than the Gaussian distribution and also exhibit asymmetry.

The four parameters of the q-distribution model explicitly represent the location, scaling parameters of positive and negative deviations and heaviness of tails. When one or two parameters reach their boundary, e.g. zero or infinity, the limiting distributions can be identical to one of the classical distributions. The Gaussian, Laplace, logistic, and exponential are among the known special cases,[2] which have numerous applications in many scientific fields[which?].

Definition

[edit]

The q-distribution has a probability density function [1] f(x)=1C(α,β,θ)[(1+ex−μαθ)(1+e−x−μβθ)]−θ for x∈ℝ, where C(α,β,θ) is a normalizing constant that often needs to be computed numerically except for the symmetric cases when θ is an integer.

Density function of a symmetric q-distribution versus Gaussian distribution
Density of an asymmetric q-distribution versus Gaussian distribution

The family of q-distributions has four parameters:

α, β, and θ must be positive.

Symmetric case

[edit]

The q-distribution is symmetric when the scaling parameters α and β are identical. In this case, the probability density function is simplified to f(x)=1αθB(θ,θ)[(1+ex−μαθ)(1+e−x−μαθ)]−θ,

where the scaling parameters are equal and the normalizing constant can be written as the Beta function. When the kurtosis parameter θ is an integer, the normalizing constant can be easily calculated. For instance, when θ=1 and B(1,1)=1, one gets the logistic distribution. The symmetric q-distribution can be directly derived from Beta distribution [2] though variable exchange with a logistic function.

The q2 distribution

[edit]

A special case of the q-distribution is when θ=2. In this case, we call it a q2 distribution. It mimics the Gaussian distribution in many ways. f(x)=3α[(1+ex−μ2α)(1+e−x−μ2α)]−2

which is symmetric with exponential tails. Its MLE estimator is similar to Huber's [3] M-estimation, which is known to be robust and insensitive to outlier data points.

Other limiting cases

[edit]

A few classical distributions can be derived from the q-distribution. These are the boundary cases when one or more of the parameters reaches zero or infinity.

f(x)→12πσe−(x−μ)22σ2
f(x)→1α[(1+e(x−μ)α)(1+e−(x−μ)α)]−1
When α≠β, it defines an asymmetric logistic distribution[4].
f(x)→1α+βe−(x−μ)+α−(x−μ)−β
This is different from the definition from Kotz et al. [5] When α=β, it becomes the Laplace distribution.
f(x)→1αe−(x−μ)+α

Properties

[edit]

The q-distribution has the following known properties:

  • Skewness ranges from −2 to 2.
  • Kurtosis ranges from 3 to 9.
  • Finite moments for any n>0
E|X|n<∞.
  • Finite expectation for λ<α−1
E(eλX)<∞.

Computational method

[edit]

With a data sample {x1,x2,…,xn}, the parameters of q-distribution can be estimated using the maximum likelihood method.

In general, for asymmetric q-distribution, there is no closed-form expression for the cumulative distribution function or the compensator C(α,β,θ). However, expectation related to q-distribution can be computed effectively using a Gaussian quadrature approach since the distribution has exponential tails.

For any smooth function, the following integral can be effectively computed using Gauss–Laguerre quadrature[6] method: ∫0∞e−xG(x)dx=∑i=1kwiG(xi) where {xi,wi} are the quadrature abscissas and weights.

Any expectation with respect to q-density can be written into the above format via variable exchanges [1] This makes it convenient to compute the compensator C(α,β,θ) as well as the cumulative distribution function and any other types of expected values.

Applications

[edit]

For applications where a Gaussian distribution is used, but is not perfect because the data set presents either asymmetry or long tails, the q-distribution can jump in as a helper.

The four parameters of the q-distribution have the flexibility to match moments up to the fourth order instead of the second order as with the Gaussian.  The distribution has potential applications in risk management, e.g. projection of tail risk, and many other scientific fields. [citation needed]

See also

[edit]

The following classical distributions are related to the q-distribution.

References

[edit]
  1. 1 2 3 Zhao, Q (2024). "A Leptokurtic Distribution Explains Volatility Skew and Smile". Journal of Derivatives. 32: 81–108. doi:10.3905/jod.2024.1.211.
  2. 1 2 Johnson N. L. and S. Lotz. (1970). Continuous univariate distributions I–II. John Wiley and Sons.
  3. ↑ Huber, P. J. (1964). "Robust Estimation of a Location Parameter". Annals of Mathematical Statistics. 35: 73–101.
  4. ↑ Stukel, T. (1988). "Generalized logistic models". Journal of the American Statistical Association. 83: 426–431.
  5. ↑ Kotz, S.; Kozubowski, T. J.; Podgorski, K. (2002). "Maximum likelihood estimation of asymmetric Laplace parameters". Annals of the Institute of Statistical Mathematics. 54: 816–826.
  6. ↑ Press, W. H.; Teukolsky, S. A.; Vetterling, W. T.; Flannery, B. P. (1992). Numerical Recipes in C. Cambridge University Press.

Category:Probability distributions