en>Gareth Jones: use template for Le Boudec reference

2011-10-06T12:16:03Z

use template for Le Boudec reference

New page

In [[multivariate statistics]], a '''data matrix''' is a mathematical [[Matrix (mathematics)|matrix]] of [[data]] of dimension ''n''-by-''p'', where ''n'' is the number of samples drawn, and ''p'' is the number of variables in each sample.<ref>{{cite web|title=Basic Concepts for Multivariate Statistics p.2|url=http://support.sas.com/publishing/pubcat/chaps/56902.pdf|publisher=SAS Institute}}</ref>

In this representation different rows represent different repetitions of an experiment, while columns represent different types of data (say, the results from particular probes). For example, suppose an experiment is run where 10 people are pulled off the street and asked four questions. The data matrix ''M'' would be a 10×4 matrix (meaning 10 rows and 4 columns). The datum in row ''i'' and column ''j'' of this matrix would be the answer of the ''i'' <sup>th</sup> person to the ''j'' <sup>th</sup> question.

==Related topics==

This arrangement reflects the convention most commonly used to present the [[linear regression]] problem, as requiring the estimation of '''B''' in the equation

:<math>\mathbf{Y} \approx \mathbf{X} \mathbf{B} </math>

where '''Y''' is an ''n''-by-''q'' data matrix containing the observed [[dependent variable]]s, '''X''' is the [[design matrix]], an ''n''-by-''p'' data matrix containing the values of the [[explanatory variable]]s, and '''B''' is a ''p''-by-''q'' matrix of unknown [[regression coefficient]]s.

==References==
{{reflist}}

[[Category:Multivariate statistics]]

{{statistics-stub}}

Palm calculus - Revision history

en>Gareth Jones: use template for Le Boudec reference