Skewness

In probability theory and statistics, skewness is a measure of the asymmetry of the probability distribution of a real-valued random variable about its mean. The skewness value can be positive or negative, or even undefined.

The qualitative interpretation of the skew is complicated. For a unimodal distribution, negative skew indicates that the tail on the left side of the probability density function is longer or fatter than the right side – it does not distinguish these shapes. Conversely, positive skew indicates that the tail on the right side is longer or fatter than the left side. In cases where one tail is long but the other tail is fat, skewness does not obey a simple rule. For example, a zero value indicates that the tails on both sides of the mean balance out, which is the case both for a symmetric distribution, and for asymmetric distributions where the asymmetries even out, such as one tail being long but thin, and the other being short but fat. Further, in multimodal distributions and discrete distributions, skewness is also difficult to interpret. Importantly, the skewness does not determine the relationship of mean and median.

Introduction

Consider the distribution in the figure. The bars on the right side of the distribution taper differently than the bars on the left side. These tapering sides are called tails, and they provide a visual means for determining which of the two kinds of skewness a distribution has:

1. negative skew: The left tail is longer; the mass of the distribution is concentrated on the right of the figure. The distribution is said to be left-skewed, left-tailed, or skewed to the left.^[1] Example (observations): 1,1001,1002,1003.
2. positive skew: The right tail is longer; the mass of the distribution is concentrated on the left of the figure. The distribution is said to be right-skewed, right-tailed, or skewed to the right.^[1] Example (observations): 1,2,3,1000.

Relationship of mean and median

The skewness is not strictly connected with the relationship between the mean and median: a distribution with negative skew can have the mean greater than or less than the median, and likewise for positive skew.

In the older notion of nonparametric skew, defined as where µ is the mean, ν is the median, and σ is the standard deviation, the skewness is defined in terms of this relationship: positive/right nonparametric skew means the mean is greater than (to the right of) the median, while negative/left nonparametric skew means the mean is less than (to the left of) the median. However, the modern definition of skewness and the traditional nonparametric definition do not in general have the same sign: while they agree for some families of distributions, they differ in general, and conflating them is misleading.

If the distribution is symmetric then the mean is equal to the median and the distribution will have zero skewness.^[2] If, in addition, the distribution is unimodal, then the mean = median = mode. This is the case of a coin toss or the series 1,2,3,4,... Note, however, that the converse is not true in general, i.e. zero skewness does not imply that the mean is equal to the median.

"Many textbooks," a 2005 article points out, "teach a rule of thumb stating that the mean is right of the median under right skew, and left of the median under left skew. [But] this rule fails with surprising frequency. It can fail in multimodal distributions, or in distributions where one tail is long but the other is fat. Most commonly, though, the rule fails in discrete distributions where the areas to the left and right of the median are not equal. Such distributions not only contradict the textbook relationship between mean, median, and skew, they also contradict the textbook interpretation of the median."^[3]

Definition

The skewness of a random variable X is the third standardized moment, denoted γ₁ and defined as

where μ₃ is the third central moment, μ is the mean, σ is the standard deviation, and E is the expectation operator. The last equality expresses skewness in terms of the ratio of the third cumulant κ₃ and the 1.5th power of the second cumulant κ₂. This is analogous to the definition of kurtosis as the fourth cumulant normalized by the square of the second cumulant.

The skewness is also sometimes denoted Skew[X].

The formula expressing skewness in terms of the non-central moment E[X³] can be expressed by expanding the previous formula,

Sample skewness

For a sample of n values the sample skewness is

where is the sample mean, m₃ is the sample third central moment, and m₂ is the sample variance.

Given samples from a population, the equation for the sample skewness above is a biased estimator of the population skewness. (Note that for a discrete distribution the sample skewness may be undefined (0/0), so its expected value will be undefined.) The usual estimator of population skewness is^{[citation needed]}

where is the unique symmetric unbiased estimator of the third cumulant and is the symmetric unbiased estimator of the second cumulant. Unfortunately is, nevertheless, generally biased (although it obviously has the correct expected value of zero for a symmetric distribution). Its expected value can even have the opposite sign from the true skewness. For instance a mixed distribution consisting of very thin Gaussians centred at −99, 0.5, and 2 with weights 0.01, 0.66, and 0.33 has a skewness of about −9.77, but in a sample of 3, has an expected value of about 0.32, since usually all three samples are in the positive-valued part of the distribution, which is skewed the other way.

The variance of the skewness of a sample of size n from a normal distribution is^[4][5]

An approximate alternative is 6/n but this is inaccurate for small samples.

Properties

Skewness can be infinite, as when

or undefined, as when

In this latter example, the third cumulant is undefined. One can also have distributions such as

where both the second and third cumulants are infinite, so the skewness is again undefined.

If Y is the sum of n independent and identically distributed random variables, all with the distribution of X, then the third cumulant of Y is n times that of X and the second cumulant of Y is n times that of X, so . This shows that the skewness of the sum is smaller, as it approaches a Gaussian distribution in accordance with the central limit theorem.

Applications

Skewness has benefits in many areas. Many models assume normal distribution; i.e., data are symmetric about the mean. The normal distribution has a skewness of zero. But in reality, data points may not be perfectly symmetric. So, an understanding of the skewness of the dataset indicates whether deviations from the mean are going to be positive or negative.

D'Agostino's K-squared test is a goodness-of-fit normality test based on sample skewness and sample kurtosis.

In almost all countries the distribution of income is skewed to the right.

Other measures of skewness

Pearson's skewness coefficients

Karl Pearson suggested simpler calculations as a measure of skewness:^[6] the Pearson mode or first skewness coefficient,^[7] defined by

• (mean − mode) / standard deviation,

as well as Pearson's median or second skewness coefficient,^[8] defined by

• 3 (mean − median) / standard deviation.

The latter is a simple multiple of the nonparametric skew.

Starting from a standard cumulant expansion around a Normal distribution, one can actually show that skewness = 6 (mean − median) / standard deviation ( 1 + kurtosis / 8) + O(skewness²).^{[citation needed]} One should keep in mind that above given equalities often don't hold even approximately and these empirical formulas are abandoned nowadays. There is no guarantee that these will be the same sign as each other or as the ordinary definition of skewness.

The adjusted Fisher-Pearson standardized moment coefficient is the version found in Excel and several statistical packages including Minitab, SAS and SPSS.^[9] The formula for this statistic is

where n is the sample size and s is the sample standard deviation.

Quantile based measures

A skewness function

can be defined,^[10][11] where F is the cumulative distribution function. This leads to a corresponding overall measure of skewness^[10] defined as the supremum of this over the range 1/2 ≤ u < 1. Another measure can be obtained by integrating the numerator and denominator of this expression.^[12] The function γ(u) satisfies -1 ≤ γ(u) ≤ 1 and is well defined without requiring the existence of any moments of the distribution.^[12]

Galton's measure of skewness^[13] is γ(u) evaluated at u = 3 / 4. Other names for this same quantity are the Bowley Skewness,^[14] the Yule-Kendall index^[15] and the quartile skewness.

Kelley's measure of skewness uses u = 0.1.^{[citation needed]}

L-moments

Use of L-moments in place of moments provides a measure of skewness known as the L-skewness.^[16]

Cyhelský's skewness coefficient

An alternative skewness coefficient may be derived from the sample mean and the individual observations:^[17]

a = ( number of observations below the mean - number of observations above the mean ) / total number of observations

The distribution of the skewness coefficient a in large sample sizes (≥45) approaches that of a normal distribution. If the variates have a normal or a uniform distribution the distribution of a is the same. The behavior of a when the variates have other distributions is currently unknown. Although this measure of skewness is very intuitive, an analytic approach to its distribution has proven difficult.

Distance skewness

A value of skewness equal to zero does not imply that the probability distribution is symmetric. Thus there is a need for another measure of asymmetry which has this property: such a measure was introduced in 2000.^[18] It is called distance skewness and denoted by dSkew. If X is a random variable which takes values in the d-dimensional Euclidean space, X has finite expectation, X' is an independent identically distributed copy of X and denotes the norm in the Euclidean space then a simple measure of asymmetry is

dSkew (X) := 1 - E||X-X'|| / E||X + X'|| if X is not 0 with probability one,

and dSkew (X):= 0 for X = 0 (with probability 1). Distance skewness is always between 0 and 1, equals 0 if and only if X is diagonally symmetric (X and -X has the same probability distribution) and equals 1 if and only if X is a nonzero constant with probability one.^[19] Thus there is a simple consistent statistical test of diagonal symmetry based on the sample distance skewness:

dSkew_n(X):= 1- ∑_i,j ||x_i – x_j|| / ∑_i,j||x_i + x_j||.

Groeneveld & Meeden’s coefficient

Groeneveld & Meeden have suggested, as an alternative measure of skewness,^[12]

where μ is the mean, ν is the median, |…| is the absolute value and E() is the expectation operator.

See also

Wikimedia Commons has media related to Skewness (statistics).

• Bragg peak
• Shape parameters
• Skew normal distribution
• Skewness risk

Notes

References

• Johnson, NL, Kotz, S, Balakrishnan N (1994) Continuous Univariate Distributions, Vol 1, 2nd Edition Wiley ISBN 0-471-58495-9
• MacGillivray, HL (1992). "Shape properties of the g- and h- and Johnson families". Comm. Statistics — Theory and Methods 21: 1244–1250. 

External links

Wikiversity has learning materials about Skewness

• Hazewinkel, Michiel, ed. (2001), "Asymmetry coefficient", Encyclopedia of Mathematics, Springer, ISBN 978-1-55608-010-4 
• An Asymmetry Coefficient for Multivariate Distributions by Michel Petitjean
• On More Robust Estimation of Skewness and Kurtosis Comparison of skew estimators by Kim and White.
• Closed-skew Distributions — Simulation, Inversion and Parameter Estimation

v t e Statistics   Descriptive statistics Continuous data Location Mean (Arithmetic, Geometric, Harmonic) Median Mode Dispersion Range Standard deviation Coefficient of variation Percentile Interquartile range Shape Variance Skewness Kurtosis Moments L-moments Count data Index of dispersion Summary tables Grouped data Frequency distribution Contingency table Dependence Pearson product-moment correlation Rank correlation (Spearman's rho, Kendall's tau) Partial correlation Scatter plot Statistical graphics Bar chart Biplot Box plot Control chart Correlogram Fan chart Forest plot Histogram Pie chart Q–Q plot Run chart Scatter plot Stemplot Radar chart   Data collection Designing studies Effect size Standard error Statistical power Sample size determination Survey methodology Sampling Stratified sampling Cluster sampling Opinion poll Questionnaire Controlled experiment Design of experiments Randomized experiment Random assignment Replication Blocking Factorial experiment Optimal design Uncontrolled studies Natural experiment Quasi-experiment Observational study   Statistical inference Statistical theory Sampling distribution Order statistic Scan statistic Record value Sufficiency Completeness Exponential family Permutation test (Randomization test) Empirical distribution Bootstrap U statistic Efficiency Asymptotics Robustness Frequentist inference Unbiased estimator (Mean unbiased minimum variance, Median unbiased) Biased estimators (Maximum likelihood, Method of moments, Minimum distance, Density estimation) Confidence interval Testing hypotheses Power Parametric tests (Likelihood-ratio, Wald, Score) Specific tests Z (normal) Student's t-test F Goodness of fit (Chi-squared, G, Sample source, sample normality, Skewness & kurtosis Normality, Model comparison, Model quality) Signed-rank (1-sample, 2-sample, 1-way anova) Shapiro–Wilk Kolmogorov–Smirnov Bayesian inference Bayesian probability Prior Posterior Credible interval Bayes factor Bayesian estimator Maximum posterior estimator   Correlation and regression analysis Correlation Pearson product–moment correlation Partial correlation Confounding variable Coefficient of determination Regression analysis Errors and residuals Regression model validation Mixed effects models Simultaneous equations models MARS Linear regression Simple linear regression Ordinary least squares General linear model Bayesian regression Non-standard predictors Nonlinear regression Nonparametric Semiparametric Isotonic Robust Heteroscedasticity Homoscedasticity Generalized linear model Exponential families Logistic (Bernoulli) Binomial Poisson Partition of variance Analysis of variance (ANOVA) Analysis of covariance Multivariate ANOVA Degrees of freedom   Categorical, multivariate, time-series, or survival analysis Categorical data Cohen's kappa Contingency table Graphical model Log-linear model McNemar's test Multivariate statistics Multivariate regression Principal components Factor analysis Cluster analysis Classification Copulas Time series analysis General Decomposition Trend Stationarity Seasonal adjustment Exponential smoothing Cointegration Specific tests Granger causality Q-Statistic Durbin–Watson Time domain ACF PACF XCF ARMA model ARIMA model ARCH Vector autoregression Frequency domain Spectral density estimation Fourier analysis Survival analysis Survival function Kaplan–Meier Logrank test Failure rate Proportional hazards models Accelerated failure time model   Applications Biostatistics Bioinformatics Clinical trials & studies Epidemiology Medical statistics Engineering statistics Chemometrics Methods engineering Probabilistic design Process & Quality control Reliability System identification Social statistics Actuarial science Census Crime statistics Demography Econometrics National accounts Official statistics Population Psychometrics Spatial statistics Cartography Environmental statistics Geographic information system Geostatistics Kriging Category Portal Outline Index