World Library  
Flag as Inappropriate
Email this Article

Random vector

Article Id: WHEBN0000049822
Reproduction Date:

Title: Random vector  
Author: World Heritage Encyclopedia
Language: English
Subject: Random variable, Random element, Vector (mathematics and physics)
Collection:
Publisher: World Heritage Encyclopedia
Publication
Date:
 

Random vector

In mathematics, probability, and statistics, a multivariate random variable or random vector is a list of mathematical variables each of whose value is unknown, either because the value has not yet occurred or because there is imperfect knowledge of its value. The individual variables in a random vector are grouped together because there may be correlations among them — often they represent different properties of an individual statistical unit (e.g. a particular person, event, etc.). Normally each element of a random vector is a real number.

Random vectors are often used as the underlying implementation of various types of aggregate random variables, e.g. a random matrix, random tree, random sequence, random process, etc.

More formally, a multivariate random variable is a column vector X = (X1, ..., Xn)T (or its transpose, which is a row vector) whose components are scalar-valued random variables on the same probability space (Ω, \mathcal{F}, P), where Ω is the sample space, \mathcal{F} is the sigma-algebra (the collection of all events), and P is the probability measure (a function returning every event's probability).

Probability distribution

Every random vector gives rise to a probability measure on Rn with the Borel algebra as the underlying sigma-algebra. This measure is also known as the joint probability distribution, the joint distribution, or the multivariate distribution of the random vector.

The distributions of each of the component random variables Xi are called marginal distributions. The conditional probability distribution of Xi given Xj is the probability distribution of Xi when Xj is known to be a particular value.

Operations on random vectors

Random vectors can be subjected to the same kinds of algebraic operations as can non-random vectors: addition, subtraction, multiplication by a scalar, and the taking of inner products.

Similarly, a new random vector Y can be defined by applying an affine transformation g\colon \mathbb{R}^n \rightarrow \mathbb{R}^n to a random vector X:

Y=\mathcal{A}X+b, where \mathcal{A} is a n \times n matrix and b is a n \times 1 column vector.

If \mathcal{A} is invertible and the probability density of X is f_X, then the probability density of Y is

f_Y(y)=\frac{f_X(\mathcal{A}^{-1}(y-b))}{|\det\mathcal{A}|}.

Expected value, covariance, and cross-covariance

The expected value or mean of a random vector X is a fixed vector E(X) whose elements are the expected values of the respective random variables.

The covariance matrix (also called the variance-covariance matrix) of an n × 1 random vector is an n × n matrix whose i, j element is the covariance between the ith and the jth random variables. The covariance matrix is the expected value, element by element, of the n × n matrix computed as [X – E(X)][X-E(X)]T, where the superscript T refers to the transpose of the indicated vector:

\operatorname{Var}(X)=\operatorname{E}[(X-\operatorname{E}[X])(X-\operatorname{E}[X])^{T}].

By extension, the cross-covariance matrix between two random vectors X and Y (X having n elements and Y having p elements) is the n × p matrix

\operatorname{Cov}(X,Y)=\operatorname{E}[(X-\operatorname{E}[X])(Y-\operatorname{E}[Y])^{T}],

where again the indicated matrix expectation is taken element-by-element in the matrix. The cross-covariance matrix Cov(Y, X) is simply the transpose of the matrix Cov(X, Y).

Further properties

Expectation of a quadratic form

One can take the expectation of a quadratic form in the random vector X as follows:[1]:p.170–171

\operatorname{E}(X^{T}AX) = [\operatorname{E}(X)]^{T}A[\operatorname{E}(X)] + \operatorname{tr}(AC),

where C is the covariance matrix of X and tr refers to the trace of a matrix — that is, to the sum of the elements on its main diagonal (from upper left to lower right). Since the quadratic form is a scalar, so is its expectation.

Proof: Let z be an m × 1 random vector with E(z) = μ and Cov(z) = V and let A be an m × m non-stochastic matrix.

Based on the formula of the covariance, then if we call z′ = X and z′A′ = Y, we see that:

\operatorname{Cov}(X,Y) = E(XY')-E(X)E(Y)'

Hence

\begin{align}

E(XY') &= \operatorname{Cov}(X,Y)+E(X)E(Y)' \\ E(z'Az) &= \operatorname{Cov}(z',z'A')+E(z')E(z'A')' \\ &=\operatorname{Cov}(z', z'A') + \mu' (\mu'A')' \\ &=\operatorname{Cov}(z', z'A') + \mu' A \mu , \end{align}

which leaves us to show that

\operatorname{Cov}(z', z'A')=\operatorname{t}(AV).

This is true based on the fact that one can cyclically permute matrices when taking a trace without changing the end result (e.g.: trace(AB) = trace(BA)).

We see that

\begin{align}

\operatorname{Cov}(z',z'A') &= E\left[\left(z' - E(z') \right)\left(z'A' - E\left(z'A'\right) \right)' \right] \\ &= E\left[ (z' - \mu') (z'A' - \mu' A' )' \right]\\ &= E\left[ (z - \mu)' (Az - A\mu) \right]. \end{align}

And since

\left( {z - \mu } \right)'\left( {Az - A\mu } \right)

is a fixed number, then

(z - \mu)' ( Az - A\mu)= \operatorname{trace}\left[ {(z - \mu )'(Az - A\mu )} \right] = \operatorname{trace} \left[(z - \mu )'A(z - \mu ) \right]

trivially. Using the permutation we get:

\operatorname{trace}\left[ {(z - \mu )'A(z - \mu )} \right] = \operatorname{trace}\left[ {A(z - \mu )'(z - \mu )} \right],

and by plugging this into the original formula we get:

\begin{align}

\operatorname{Cov} \left( {z',z'A'} \right) &= E\left[ {\left( {z - \mu } \right)' (Az - A\mu)} \right] \\ &= E \left[ \operatorname{trace}\left[ A(z - \mu )'(z - \mu )\right] \right] \\ &= \operatorname{trace} \left[ {A \cdot E \left[(z - \mu )'(z - \mu )\right] } \right] \\ &= \operatorname{trace} [A V]. \end{align}

Expectation of the product of two different quadratic forms

One can take the expectation of the product of two different quadratic forms in a zero-mean Gaussian random vector X as follows:[1]:pp. 162–176

\operatorname{E}[X^{T}AX][X^{T}BX] = 2\operatorname{trace}(ACBC) + \operatorname{trace}(AC)\operatorname{trace}(BC)

where again C is the covariance matrix of X. Again, since both quadratic forms are scalars and hence their product is a scalar, the expectation of their product is also a scalar.

Applications

Portfolio theory

In portfolio theory in finance, an objective often is to choose a portfolio of risky assets such that the distribution of the random portfolio return has desirable properties. For example, one might want to choose the portfolio return having the lowest variance for a given expected value. Here the random vector is the vector r of random returns on the individual assets, and the portfolio return p (a random scalar) is the inner product of the vector of random returns with a vector w of portfolio weights — the fractions of the portfolio placed in the respective assets. Since p = wTr, the expected value of the portfolio return is wTE(r) and the variance of the portfolio return can be shown to be wTCw, where C is the covariance matrix of r.

Regression theory

In linear regression theory, we have data on n observations on a dependent variable y and n observations on each of k independent variables xj. The observations on the dependent variable are stacked into a column vector y; the observations on each independent variable are also stacked into column vectors, and these latter column vectors are combined into a matrix X of observations on the independent variables. Then the following regression equation is postulated as a description of the process that generated the data:

y = X \beta + e,

where β is a postulated fixed but unknown vector of k response coefficients, and e is an unknown random vector reflecting random influences on the dependent variable. By some chosen technique such as ordinary least squares, a vector \hat \beta is chosen as an estimate of β, and the estimate of the vector e, denoted \hat e, is computed as

\hat e = y - X \hat \beta.

Then the statistician must analyze the properties of \hat \beta and \hat e, which are viewed as random vectors since a randomly different selection of n cases to observe would have resulted in different values for them.

References

de:Zufallsvariable#Mehrdimensionale Zufallsvariable

pl:Zmienna losowa#Uogólnienia

This article was sourced from Creative Commons Attribution-ShareAlike License; additional terms may apply. World Heritage Encyclopedia content is assembled from numerous content providers, Open Access Publishing, and in compliance with The Fair Access to Science and Technology Research Act (FASTR), Wikimedia Foundation, Inc., Public Library of Science, The Encyclopedia of Life, Open Book Publishers (OBP), PubMed, U.S. National Library of Medicine, National Center for Biotechnology Information, U.S. National Library of Medicine, National Institutes of Health (NIH), U.S. Department of Health & Human Services, and USA.gov, which sources content from all federal, state, local, tribal, and territorial government publication portals (.gov, .mil, .edu). Funding for USA.gov and content contributors is made possible from the U.S. Congress, E-Government Act of 2002.
 
Crowd sourced content that is contributed to World Heritage Encyclopedia is peer reviewed and edited by our editorial staff to ensure quality scholarly research articles.
 
By using this site, you agree to the Terms of Use and Privacy Policy. World Heritage Encyclopedia™ is a registered trademark of the World Public Library Association, a non-profit organization.
 


Copyright © World Library Foundation. All rights reserved. eBooks from Project Gutenberg are sponsored by the World Library Foundation,
a 501c(4) Member's Support Non-Profit Organization, and is NOT affiliated with any governmental agency or department.