Location–scale family explained

X

whose probability distribution function belongs to such a family, the distribution function of

Y\stackrel{d}{=}a+bX

also belongs to the family (where

\stackrel{d}{=}

means "equal in distribution"—that is, "has the same distribution as").

In other words, a class

\Omega

of probability distributions is a location–scale family if for all cumulative distribution functions

F\in\Omega

and any real numbers

a\inR

and

b>0

, the distribution function

G(x)=F(a+bx)

is also a member of

\Omega

.

X

has a cumulative distribution function

FX(x)=P(X\lex)

, then

Y{=}a+bX

has a cumulative distribution function

FY(y)=

F
X\left(y-a
b

\right)

.

X

is a discrete random variable with probability mass function

pX(x)=P(X=x)

, then

Y{=}a+bX

is a discrete random variable with probability mass function

pY(y)=

p
X\left(y-a
b

\right)

.

X

is a continuous random variable with probability density function

fX(x)

, then

Y{=}a+bX

is a continuous random variable with probability density function

fY(y)=

1
b
f
X\left(y-a
b

\right)

.

Moreover, if

X

and

Y

are two random variables whose distribution functions are members of the family, and assuming existence of the first two moments and

X

has zero mean and unit variance,then

Y

can be written as

Y\stackrel{d}{=}\muY+\sigmaYX

, where

\muY

and

\sigmaY

are the mean and standard deviation of

Y

.

In decision theory, if all alternative distributions available to a decision-maker are in the same location–scale family, and the first two moments are finite, then a two-moment decision model can apply, and decision-making can be framed in terms of the means and the variances of the distributions.[1] [2] [3]

Examples

Often, location–scale families are restricted to those where all members have the same functional form. Most location–scale families are univariate, though not all. Well-known families in which the functional form of the distribution is consistent throughout the family include the following:

Converting a single distribution to a location–scale family

The following shows how to implement a location–scale family in a statistical package or programming environment where only functions for the "standard" version of a distribution are available. It is designed for R but should generalize to any language and library.

The example here is of the Student's t-distribution, which is normally provided in R only in its standard form, with a single degrees of freedom parameter df. The versions below with _ls appended show how to generalize this to a generalized Student's t-distribution with an arbitrary location parameter m and scale parameter s.

Probability density function (PDF):dt_ls(x, df, m, s) =1/s * dt((x - m) / s, df)
Cumulative distribution function (CDF):pt_ls(x, df, m, s) =pt((x - m) / s, df)
Quantile function (inverse CDF):qt_ls(prob, df, m, s) =qt(prob, df) * s + m
Generate a random variate:rt_ls(df, m, s) =rt(df) * s + m
Note that the generalized functions do not have standard deviation s since the standard t distribution does not have standard deviation of 1.

External links

Notes and References

  1. Meyer . Jack . Two-Moment Decision Models and Expected Utility Maximization . . 77 . 1987 . 3 . 421–430 . 1804104 .
  2. Mayshar . J. . A Note on Feldstein's Criticism of Mean-Variance Analysis . . 45 . 1 . 1978 . 197–199 . 2297094 .
  3. Book: Sinn, H.-W. . Hans-Werner Sinn . Economic Decisions under Uncertainty . Second English . 1983 . North-Holland .