Range (statistics)

Not to be confused with Mid-range.

For other uses, see Range (disambiguation) § Mathematics.

In descriptive statistics, the range of a set of data is size of the narrowest interval which contains all the data. It is calculated as the difference between the largest and smallest values, ^[1] the result of subtracting the sample maximum and minimum. It is expressed in the same units as the data. The range provides an indication of statistical dispersion. Since it only depends on two of the observations, it is most useful in representing the dispersion of small data sets. ^[2]

For continuous IID random variables

For n independent and identically distributed continuous random variables X₁, X₂, ..., X_n with the cumulative distribution function G(x) and a probability density function g(x), let T denote the range of them, that is, T= max(X₁, X₂, ..., X_n)- min(X₁, X₂, ..., X_n).

Distribution

The range, T, has the cumulative distribution function ^{[3] [4]}

${\displaystyle F(t)=n\int _{-\infty }^{\infty }g(x)[G(x+t)-G(x)]^{n-1}\,{\text{d}}x.}$

Gumbel notes that the "beauty of this formula is completely marred by the facts that, in general, we cannot express G(x + t) by G(x), and that the numerical integration is lengthy and tiresome." ^{[3] : 385}

If the distribution of each X_i is limited to the right (or left) then the asymptotic distribution of the range is equal to the asymptotic distribution of the largest (smallest) value. For more general distributions the asymptotic distribution can be expressed as a Bessel function. ^[3]

Moments

The mean range is given by ^[5]

${\displaystyle n\int _{0}^{1}x(G)[G^{n-1}-(1-G)^{n-1}]\,{\text{d}}G}$

where x(G) is the inverse function. In the case where each of the X_i has a standard normal distribution, the mean range is given by ^[6]

${\displaystyle \int _{-\infty }^{\infty }(1-(1-\Phi (x))^{n}-\Phi (x)^{n})\,{\text{d}}x.}$

For continuous non-IID random variables

For n nonidentically distributed independent continuous random variables X₁, X₂, ..., X_n with cumulative distribution functions G₁(x), G₂(x), ..., G_n(x) and probability density functions g₁(x), g₂(x), ..., g_n(x), the range has cumulative distribution function ^[4]

${\displaystyle F(t)=\sum _{i=1}^{n}\int _{-\infty }^{\infty }g_{i}(x)\prod _{j=1,j\neq i}^{n}[G_{j}(x+t)-G_{j}(x)]\,{\text{d}}x.}$

For discrete IID random variables

For n independent and identically distributed discrete random variables X₁, X₂, ..., X_n with cumulative distribution function G(x) and probability mass function g(x) the range of the X_i is the range of a sample of size n from a population with distribution function G(x). We can assume without loss of generality that the support of each X_i is {1,2,3,...,N} where N is a positive integer or infinity. ^{[7] [8]}

Distribution

The range has probability mass function ^{[7] [9] [10]}

${\displaystyle f(t)={\begin{cases}\sum _{x=1}^{N}[g(x)]^{n}&t=0\\[6pt]\sum _{x=1}^{N-t}\left({\begin{alignedat}{2}&[G(x+t)-G(x-1)]^{n}\\{}-{}&[G(x+t)-G(x)]^{n}\\{}-{}&[G(x+t-1)-G(x-1)]^{n}\\{}+{}&[G(x+t-1)-G(x)]^{n}\\\end{alignedat}}\right)&t=1,2,3\ldots ,N-1.\end{cases}}}$

Example

If we suppose that g(x) = 1/N, the discrete uniform distribution for all x, then we find ^{[9] [11]}

${\displaystyle f(t)={\begin{cases}{\frac {1}{N^{n-1}}}&t=0\\[4pt]\sum _{x=1}^{N-t}\left(\left[{\frac {t+1}{N}}\right]^{n}-2\left[{\frac {t}{N}}\right]^{n}+\left[{\frac {t-1}{N}}\right]^{n}\right)&t=1,2,3\ldots ,N-1.\end{cases}}}$

Derivation

The probability of having a specific range value, t, can be determined by adding the probabilities of having two samples differing by t, and every other sample having a value between the two extremes. The probability of one sample having a value of x is ${\displaystyle ng(x)}$. The probability of another having a value t greater than x is:

${\displaystyle (n-1)g(x+t).}$

The probability of all other values lying between these two extremes is:

${\displaystyle \left(\int _{x}^{x+t}g(x)\,{\text{d}}x\right)^{n-2}=\left(G(x+t)-G(x)\right)^{n-2}.}$

Combining the three together yields:

${\displaystyle f(t)=n(n-1)\int _{-\infty }^{\infty }g(x)g(x+t)[G(x+t)-G(x)]^{n-2}\,{\text{d}}x}$

Related quantities

The range is a specific example of order statistics. In particular, the range is a linear function of order statistics, which brings it into the scope of L-estimation.

See also

• Mathematics portal

• Interdecile range
• Interquartile range
• Studentized range

References

1. George Woodbury (2001). An Introduction to Statistics. Cengage Learning. p. 74. ISBN   0534377556.
2. Carin Viljoen (2000). Elementary Statistics: Vol 2. Pearson South Africa. pp. 7–27. ISBN   186891075X.
3. E. J. Gumbel (1947). "The Distribution of the Range". The Annals of Mathematical Statistics. 18 (3): 384–412. doi: 10.1214/aoms/1177730387 . JSTOR   2235736.
4. Tsimashenka, I.; Knottenbelt, W.; Harrison, P. (2012). "Controlling Variability in Split-Merge Systems". Analytical and Stochastic Modeling Techniques and Applications (PDF). Lecture Notes in Computer Science. Vol. 7314. p. 165. doi:10.1007/978-3-642-30782-9_12. ISBN   978-3-642-30781-2.
5. H. O. Hartley; H. A. David (1954). "Universal Bounds for Mean Range and Extreme Observation". The Annals of Mathematical Statistics. 25 (1): 85–99. doi: 10.1214/aoms/1177728848 . JSTOR   2236514.
6. L. H. C. Tippett (1925). "On the Extreme Individuals and the Range of Samples Taken from a Normal Population". Biometrika. 17 (3/4): 364–387. doi:10.1093/biomet/17.3-4.364. JSTOR   2332087.
7. Evans, D. L.; Leemis, L. M.; Drew, J. H. (2006). "The Distribution of Order Statistics for Discrete Random Variables with Applications to Bootstrapping". INFORMS Journal on Computing. 18: 19. doi:10.1287/ijoc.1040.0105.
8. Irving W. Burr (1955). "Calculation of Exact Sampling Distribution of Ranges from a Discrete Population". The Annals of Mathematical Statistics. 26 (3): 530–532. doi: 10.1214/aoms/1177728500 . JSTOR   2236482.
9. Abdel-Aty, S. H. (1954). "Ordered variables in discontinuous distributions". Statistica Neerlandica. 8 (2): 61–82. doi:10.1111/j.1467-9574.1954.tb00442.x.
10. Siotani, M. (1956). "Order statistics for discrete case with a numerical application to the binomial distribution". Annals of the Institute of Statistical Mathematics. 8: 95–96. doi:10.1007/BF02863574.
11. Paul R. Rider (1951). "The Distribution of the Range in Samples from a Discrete Rectangular Population". Journal of the American Statistical Association . 46 (255): 375–378. doi:10.1080/01621459.1951.10500796. JSTOR   2280515.