Skew normal distribution

Skew Normal

Probability density function
Cumulative distribution function
Parameters	$\xi \,$ location (real) $\omega \,$ scale (positive, real) $\alpha \,$ shape (real)
Support	$x\in (-\infty ;+\infty )\!$
PDF	${\frac {1}{\omega \pi }}e^{-{\frac {(x-\xi )^{2}}{2\omega ^{2}}}}\int _{-\infty }^{\alpha \left({\frac {x-\xi }{\omega }}\right)}e^{-{\frac {t^{2}}{2}}}\ dt$
CDF	$\Phi \left({\frac {x-\xi }{\omega }}\right)-2T\left({\frac {x-\xi }{\omega }},\alpha \right)$ $T(h,a)$ is Owen's T function
Mean	$\xi +\omega \delta {\sqrt {\frac {2}{\pi }}}$ where $\delta ={\frac {\alpha }{\sqrt {1+\alpha ^{2}}}}$
Variance	$\omega ^{2}\left(1-{\frac {2\delta ^{2}}{\pi }}\right)$
Skewness	${\frac {4-\pi }{2}}{\frac {\left(\delta {\sqrt {2/\pi }}\right)^{3}}{\left(1-2\delta ^{2}/\pi \right)^{3/2}}}$
Excess kurtosis	$2(\pi -3){\frac {\left(\delta {\sqrt {2/\pi }}\right)^{4}}{\left(1-2\delta ^{2}/\pi \right)^{2}}}$
MGF	$M_{X}\left(t\right)=2\exp \left(\xi t+{\frac {\omega ^{2}t^{2}}{2}}\right)\Phi \left(\omega \delta t\right)$

In probability theory and statistics, the skew normal distribution is a continuous probability distribution that generalises the normal distribution to allow for non-zero skewness.

Definition

Let $\phi (x)$ denote the standard normal probability density function

$\phi (x)={\frac {1}{\sqrt {2\pi }}}e^{-{\frac {x^{2}}{2}}}$

with the cumulative distribution function given by

$\Phi (x)=\int _{-\infty }^{x}\phi (t)\ dt={\frac {1}{2}}\left[1+{\text{erf}}\left({\frac {x}{\sqrt {2}}}\right)\right].$

Then the probability density function of the skew-normal distribution with parameter α is given by

$f(x)=2\phi (x)\Phi (\alpha x).\,$

This distribution was first introduced by O'Hagan and Leonhard (1976).

To add location and scale parameters to this, one makes the usual transform $x\rightarrow {\frac {x-\xi }{\omega }}$. One can verify that the normal distribution is recovered when $\alpha =0$, and that the absolute value of the skewness increases as the absolute value of $\alpha$ increases. The distribution is right skewed if $\alpha >0$ and is left skewed if $\alpha <0$. The probability density function with location $\xi$, scale $\omega$, and parameter $\alpha$ becomes

$f(x)={\frac {2}{\omega }}\phi \left({\frac {x-\xi }{\omega }}\right)\Phi \left(\alpha \left({\frac {x-\xi }{\omega }}\right)\right).\,$

Estimation

Maximum likelihood estimates for $\xi$, $\omega$, and $\alpha$ can be computed numerically, but no closed-form expression for the estimates is available unless $\alpha =0$. If a closed-form expression is needed, the method of moments can be applied to estimate $\alpha$ from the sample skew, by inverting the skewness equation. This yields the estimate

$|\delta |={\sqrt {{\frac {\pi }{2}}{\frac {|{\hat {\gamma }}_{3}|^{\frac {2}{3}}}{|{\hat {\gamma }}_{3}|^{\frac {2}{3}}+((4-\pi )/2)^{\frac {2}{3}}}}}}$

where $\delta ={\frac {\alpha }{\sqrt {1+\alpha ^{2}}}}$, and ${\hat {\gamma }}_{3}$ is the sample skew. The sign of $\delta$ is the same as the sign of ${\hat {\gamma }}_{3}$. Consequently, ${\hat {\alpha }}=\delta /{\sqrt {1-\delta ^{2}}}$.

The maximum (theoretical) skewness is obtained by setting ${\delta =1}$ in the skewness equation, giving $\gamma _{3}\approx 0.9952717$. However it is possible that the sample skewness is larger, and then $\alpha$ cannot be determined from these equations. When using the method of moments in an automatic fashion, for example to give starting values for maximum likelihood iteration, one should therefore set (for example) ${\hat {\gamma }}_{3}=\min(0.99,(1/n)\sum {(x_{i}-{\bar {x}})^{3}}$.

See also

• Normal distribution
• Generalized normal distribution
• Lognormal distribution
• Skewness

References

• Azzalini, A. (1985). "A class of distributions which includes the normal ones". Scand. J. Statist. 12: 171–178. {{cite journal}}: Cite has empty unknown parameter: |coauthors= (help)

• O'Hagan, A. and Leonhard, T. (1976). Bayes estimation subject to uncertainty about parameter constraints. Biometrika, 63, 201-202.

External links

• A very brief introduction to the skew-normal distribution
• The Skew-Normal Probability Distribution (and related distributions, such as the skew-t)
• OWENS: Owen's T Function
• Closed-skew Distributions - Simulation, Inversion and Parameter Estimation