Principal component analysis

9/13/2010 Principal component analysis - Wikipe…
Principal component analysis

From Wikipedia, the free encyclopedia
Principal component analysis (PCA) involves a mathematical

procedure that transforms a number of possibly correlated variables into
a smaller number of uncorrelated variables called principal components.
The first principal component accounts for as much of the variability in the
data as possible, and each succeeding component accounts for as much
of the remaining variability as possible. Depending on the field of
application, it is also named the discrete Karhunen–Loève transform
(KLT), the Hotelling transform or proper orthogonal decomposition
(POD).
PCA was invented in 1901 by Karl Pearson.[1] Now it is mostly used as

a tool in exploratory data analysis and for making predictive models. PCA of a multivariate Gaussian
PCA involves the calculation of the eigenvalue decomposition of a data distribution centered at (1,3) with a
covariance matrix or singular value decomposition of a data matrix, standard deviation of 3 in roughly the
usually after mean centering the data for each attribute. The results of a (0.878, 0.478) direction and of 1 in
PCA are usually discussed in terms of component scores and loadings the orthogonal direction.
(Shaw, 2003)[citation needed].
PCA is the simplest of the true eigenvector-based multivariate analyses. Often, its operation can be thought of as
revealing the internal structure of the data in a way which best explains the variance in the data. If a multivariate
dataset is visualised as a set of coordinates in a high-dimensional data space (1 axis per variable), PCA supplies the
user with a lower-dimensional picture, a "shadow" of this object when viewed from its (in some sense) most
informative viewpoint.
PCA is closely related to factor analysis; indeed, some statistical packages deliberately conflate the two techniques.
True factor analysis makes different assumptions about the underlying structure and solves eigenvectors of a slightly
different matrix.
Contents
1 Details
2 Discussion
3 Table of symbols and abbreviations
4 Properties and limitations of PCA
5 Computing PCA using the covariance method
5.1 Organize the data set
5.2 Calculate the empirical mean
5.3 Calculate the deviations from the mean
5.4 Find the covariance matrix
5.5 Find the eigenvectors and eigenvalues of the covariance matrix
…wikipedia.org/…/Principal_componen… 1/15
5.6 Rearrange the eigenvectors and eigenvalues
5.7 Compute the cumulative energy content for each eigenvector
5.8 Select a subset of the eigenvectors as basis vectors
5.9 Convert the source data to z-scores
5.10 Project the z-scores of the data onto the new basis
6 Derivation of PCA using the covariance method
7 Computing principal components with expectation maximization
7.1 The NIPALS method
8 Relation between PCA and K-means clustering
9 Correspondence analysis
10 Generalizations
10.1 Nonlinear generalizations
10.2 Higher order
10.3 Robustness - Weighted PCA
11 Software/source code
12 Notes
13 See also
14 References
15 External links
15.1 Non-technical introductions
Details
PCA is mathematically defined[2] as an orthogonal linear transformation that transforms the data to a new
coordinate system such that the greatest variance by any projection of the data comes to lie on the first coordinate
(called the first principal component), the second greatest variance on the second coordinate, and so on. PCA is
theoretically the optimum transform for given data in least square terms.
For a data matrix, XT, with zero empirical mean (the empirical mean of the distribution has been subtracted from
the data set), where each row represents a different repetition of the experiment, and each column gives the results
from a particular probe, the PCA transformation is given by:
where the matrix Σ is an m-by-n diagonal matrix with nonnegative real numbers on the diagonal and W Σ VT is the
singular value decomposition (svd) of X.
Given a set of points in Euclidean space, the first principal component (the eigenvector with the largest eigenvalue)
corresponds to a line that passes through the mean and minimizes sum squared error with those points. The second
principal component corresponds to the same concept after all correlation with the first principal component has
been subtracted out from the points. Each eigenvalue indicates the portion of the variance that is correlated with
each eigenvector. Thus, the sum of all the eigenvalues is equal to the sum squared distance of the points with their
mean divided by the number of dimensions. PCA essentially rotates the set of points around their mean in order to
align with the first few principal components. This moves as much of the variance as possible (using a linear
transformation) into the first few dimensions. The values in the remaining dimensions, therefore, tend to be highly
correlated and may be dropped with minimal loss of information. PCA is often used in this manner for
dimensionality reduction. PCA has the distinction of being the optimal linear transformation for keeping the
subspace that has largest variance. This advantage, however, comes at the price of greater computational
requirement if compared, for example, to the discrete cosine transform. Nonlinear dimensionality reduction
techniques tend to be more computationally demanding than PCA.
Discussion
Mean subtraction (a.k.a. "mean centering") is necessary for performing PCA to ensure that the first principal
component describes the direction of maximum variance. If mean subtraction is not performed, the first principal
component will instead correspond to the mean of the data. A mean of zero is needed for finding a basis that
minimizes the mean square error of the approximation of the data[3].
Assuming zero empirical mean (the empirical mean of the distribution has been subtracted from the data set), the
principal component w1 of a data set x can be defined as:
(See arg max for the notation.) With the first k − 1 components, the kth component can be found by subtracting the
first k − 1 principal components from x:
and by substituting this as the new data set to find a principal component in
The Karhunen–Loève transform is therefore equivalent to finding the singular value decomposition of the data matrix
X,
and then obtaining the reduced-space data matrix Y by projecting X down into the reduced space defined by only
the first L singular vectors, WL:
The matrix W of singular vectors of X is equivalently the matrix W of eigenvectors of the matrix of observed
covariances C = X XT,
The eigenvectors with the largest eigenvalues correspond to the dimensions that have the strongest correlation in the
data set (see Rayleigh quotient).
PCA is equivalent to empirical orthogonal functions (EOF), a name which is used in meteorology.
An autoencoder neural network with a linear hidden layer is similar to PCA. Upon convergence, the weight vectors
of the K neurons in the hidden layer will form a basis for the space spanned by the first K principal components.
Unlike PCA, this technique will not necessarily produce orthogonal vectors.
PCA is a popular primary technique in pattern recognition. But it is not optimized for class separability[4]. An
alternative is the linear discriminant analysis, which does take this into account.
Table of symbols and abbreviations

Symbol Meaning Dimensions Indices
data matrix, consisting of the set of all data vectors,
one vector per column
the number of column vectors in the data set scalar
the number of elements in each column vector
scalar
(dimension)
the number of dimensions in the dimensionally reduced
scalar
subspace,
vector of empirical means, one mean for each row m
of the data matrix
vector of empirical standard deviations, one standard
deviation for each row m of the data matrix
vector of all 1's
deviations from the mean of each row m of the data
matrix
z-scores, computed using the mean and standard
deviation for each row m of the data matrix
covariance matrix
correlation matrix
matrix consisting of the set of all eigenvectors of C,

one eigenvector per column
diagonal matrix consisting of the set of all eigenvalues
of C along its principal diagonal, and 0 for all other
elements
matrix of basis vectors, one vector per column, where
each basis vector is one of the eigenvectors of C, and
where the vectors in W are a sub-set of those in V
matrix consisting of N column vectors, where each
vector is the projection of the corresponding data
vector from matrix X onto the basis vectors contained
in the columns of matrix W.
Properties and limitations of PCA

PCA is theoretically the optimal linear scheme, in terms of least mean square error, for compressing a set of high
dimensional vectors into a set of lower dimensional vectors and then reconstructing the original set. It is a non-
parametric analysis and the answer is unique and independent of any hypothesis about data probability distribution.
However, the latter two properties are regarded as weakness as well as strength, in that being non-parametric, no
prior knowledge can be incorporated and that PCA compressions often incur loss of information.
The applicability of PCA is limited by the assumptions[5] made in its derivation. These assumptions are:
Assumption on linearity
We assumed the observed data set to be linear combinations of certain basis. Non-linear methods such as
kernel PCA have been developed without assuming linearity.
Assumption on the statistical importance of mean and covariance
PCA uses the eigenvectors of the covariance matrix and it only finds the independent axes of the data under
the Gaussian assumption. For non-Gaussian or multi-modal Gaussian data, PCA simply de-correlates the
axes. When PCA is used for clustering, its main limitation is that it does not account for class separability
since it makes no use of the class label of the feature vector. There is no guarantee that the directions of
maximum variance will contain good features for discrimination.
Assumption that large variances have important dynamics
PCA simply performs a coordinate rotation that aligns the transformed axes with the directions of maximum
variance. It is only when we believe that the observed data has a high signal-to-noise ratio that the principal
components with larger variance correspond to interesting dynamics and lower ones correspond to noise.
Essentially, PCA involves only rotation and scaling. The above assumptions are made in order to simplify the
algebraic computation on the data set. Some other methods have been developed without one or more of these
assumptions; these are described below.
Computing PCA using the covariance method

Following is a detailed description of PCA using the covariance method (see also here
(https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.cs.otago.ac.nz%2Fcosc453%2Fstudent_tutorials%2Fprincipal_components.pdf) ).
The goal is to transform a given data set X of dimension M to an alternative data set Y of smaller dimension L.
Equivalently, we are seeking to find the matrix Y, where Y is the Karhunen–Loève transform (KLT) of matrix X:
Organize the data set
Suppose you have data comprising a set of observations of M variables, and you want to reduce the data so that
each observation can be described with only L variables, L < M. Suppose further, that the data are arranged as a
set of N data vectors with each representing a single grouped observation of the M variables.
Write as column vectors, each of which has M rows.

Place the column vectors into a single matrix X of dimensions M × N.
Calculate the empirical mean
Find the empirical mean along each dimension m = 1, ..., M.

Place the calculated mean values into an empirical mean vector u of dimensions M × 1.
Calculate the deviations from the mean
Mean subtraction is an integral part of the solution towards finding a principal component basis that minimizes the
mean square error of approximating the data[6]. Hence we proceed by centering the data as follows:
Subtract the empirical mean vector u from each column of the data matrix X.
Store mean-subtracted data in the M × N matrix B.
where h is a 1 × N row vector of all 1s:
Find the covariance matrix
Find the M × M empirical covariance matrix C from the outer product of matrix B with itself:
where
is the expected value operator,

is the outer product operator, and
is the conjugate transpose operator. Note that if B consists entirely of real numbers, which is
the case in many applications, the "conjugate transpose" is the same as the regular transpose.
Please note that the information in this section is indeed a bit fuzzy. Outer products apply to vectors, for
tensor cases we should apply tensor products, but the covariance matrix in PCA, is a sum of outer products
between its sample vectors, indeed it could be represented as B.B*. See the covariance matrix sections on
the discussion page for more information.
Find the eigenvectors and eigenvalues of the covariance matrix
Compute the matrix V of eigenvectors which diagonalizes the covariance matrix C:
where D is the diagonal matrix of eigenvalues of C. This step will typically involve the use of a computer-
based algorithm for computing eigenvectors and eigenvalues. These algorithms are readily available as sub-
components of most matrix algebra systems, such as MATLAB[7][8], Mathematica[9], SciPy, IDL(Interactive
Data Language), or GNU Octave as well as OpenCV.
Matrix D will take the form of an M × M diagonal matrix, where
is the mth eigenvalue of the covariance matrix C, and
Matrix V, also of dimension M × M, contains M column vectors, each of length M, which represent the M
eigenvectors of the covariance matrix C.
The eigenvalues and eigenvectors are ordered and paired. The mth eigenvalue corresponds to the mth
eigenvector.
Rearrange the eigenvectors and eigenvalues
Sort the columns of the eigenvector matrix V and eigenvalue matrix D in order of decreasing eigenvalue.
Make sure to maintain the correct pairings between the columns in each matrix.
Compute the cumulative energy content for each eigenvector

The eigenvalues represent the distribution of the source data's energy among each of the eigenvectors, where
the eigenvectors form a basis for the data. The cumulative energy content g for the mth eigenvector is the
sum of the energy content across all of the eigenvalues from 1 through m:
Select a subset of the eigenvectors as basis vectors
Save the first L columns of V as the M × L matrix W:
where
Use the vector g as a guide in choosing an appropriate value for L. The goal is to choose a value of L as
small as possible while achieving a reasonably high value of g on a percentage basis. For example, you may
want to choose L so that the cumulative energy g is above a certain threshold, like 90 percent. In this case,
choose the smallest value of L such that
Convert the source data to z-scores
Create an M × 1 empirical standard deviation vector s from the square root of each element along the main
diagonal of the covariance matrix C:
Calculate the M × N z-score matrix:
(divide element-by-element)
Note: While this step is useful for various applications as it normalizes the data set with respect to its
variance, it is not integral part of PCA/KLT!
Project the z-scores of the data onto the new basis
The projected vectors are the columns of the matrix
W* is the conjugate transpose of the eigenvector matrix.

The columns of matrix Y represent the Karhunen–Loeve transforms (KLT) of the data vectors in the columns
of matrix X.
Derivation of PCA using the covariance method

Let X be a d-dimensional random vector expressed as column vector. Without loss of generality, assume X has
zero mean. We want to find a orthonormal transformation matrix P such that
with the constraint that
is a diagonal matrix and
By substitution, and matrix algebra, we obtain:
We now have:
Rewrite P as d column vectors, so
and as:
Substituting into equation above, we obtain:
Notice that in , Pi is an eigenvector of the covariance matrix of X. Therefore, by finding the

eigenvectors of the covariance matrix of X, we find a projection matrix P that satisfies the original constraints.
Computing principal components with expectation maximization

In practical implementations especially with high dimensional data, the covariance method is rarely used because it is
not efficient. Expectation maximization is one way to compute principal components efficiently[10]. The following
pseudo-code computes the first principal component of a data matrix, XT, with zero mean, without ever computing
its covariance[10]. Note that here a zero mean data matrix means that the columns of XT should each have zero
mean.
a random vector
do c times:
for each row
return
The algorithm is nothing but the power iteration applied to the covariance matrix expressed as a sum of outer
products (thereby avoiding direct calculation of the matrix itself, and thus requiring less computation). p will typically
converge to the first principal component of XT within a small number of iterations, c. (The magnitude of t will be
larger after each iteration. Convergence can be detected when it increases by an amount too small for the precision
of the machine.)
Subsequent principal components can be computed by subtracting component p from XT (see Gram–Schmidt) and
then repeating this algorithm to find the next principal component. However this simple approach is not numerically
stable if more than a small number of principal components are required, because imprecisions in the calculations
will additively affect the estimates of subsequent principal components. More advanced methods build on this basic
idea, as with the closely related Lanczos algorithm.
One way to compute the eigenvalue that corresponds with each principal component is to measure the difference in
sum-squared-distance between the rows and the mean, before and after subtracting out the principal component.
The eigenvalue that corresponds with the component that was removed is equal to this difference divided by the
number of dimensions (columns).
The NIPALS method

Main article: Non-linear iterative partial least squares
For very high-dimensional datasets, such as those generated in the *omics sciences (e.g., genomics, metabolomics)
it is usually only necessary to compute the first few PCs. The non-linear iterative partial least squares (NIPALS)
algorithm calculates t1 and p1' from X. The outer product, t1p1' can then be subtracted from X leaving the residual
matrix E1 . This can be then used to calculate subsequent PCs.[11] This results in a dramatic reduction in
computational time since calculation of the covariance matrix is avoided.
Relation between PCA and K-means clustering

It has been shown recently (2007) [12] [13] that the relaxed solution of K-means clustering, specified by the cluster
indicators, is given by the PCA principal components, and the PCA subspace spanned by the principal directions is
identical to the cluster centroid subspace specified by the between-class scatter matrix. Thus PCA automatically
projects to the subspace where the global solution of K-means clustering lie, and thus facilitate K-means clustering
to find near-optimal solutions.
Correspondence analysis
Correspondence analysis (CA) was developed by Jean-Paul Benzécri[14] and is conceptually similar to PCA, but
scales the data (which should be non-negative) so that rows and columns are treated equivalently. It is traditionally
applied to contingency tables. CA decomposes the chi-square statistic associated to this table into orthogonal
factors[15]. Because CA is a descriptive technique, it can be applied to tables for which the chi-square statistic is
appropriate or not. Several variants of CA are available including detrended correspondence analysis and canonical
correspondence analysis. One special extension is Multiple correspondence analysis, which may be seen as the
counterpart of principal component analysis for categorical data.[16].
Generalizations
Nonlinear generalizations
Most of the modern methods for nonlinear dimensionality

reduction find their theoretical and algorithmic roots in PCA
or K-means. Pearson's original idea was to take a straight
line (or plane) which will be "the best fit" to a set of data
points. Principal curves and manifolds [19] give the natural
geometric framework for PCA generalization and extend the
geometric interpretation of PCA by explicitly constructing an
embedded manifold for data approximation, and by
encoding using standard geometric projection onto the
manifold, as it is illustrated by Fig. See also principal
geodesic analysis.
Higher order
N-way principal component analysis may be performed with
models such as Tucker decomposition, PARAFAC,
multiple factor analysis, co-inertia analysis, STATIS, and
DISTATIS.
Robustness - Weighted PCA Linear PCA versus nonlinear Principal

Manifolds[17] for visualization of breast cancer
While PCA finds the mathematically optimal method (as in microarray data: a) Configuration of nodes and 2D
minimizing the squared error), it is sensitive to outliers in the Principal Surface in the 3D PCA linear manifold.
data that produce large errors PCA tries to avoid. It The dataset is curved and can not be mapped
therefore is common practise to remove outliers before adequately on a 2D principal plane; b) The
computing PCA. However, in some contexts, outliers can distribution in the internal 2D non-linear principal
be difficult to identify. For example in data mining algorithms surface coordinates (ELMap2D) together with an
like correlation clustering, the assignment of points to estimation of the density of points; c) The same as
clusters and outliers is not known beforehand. A recently b), but for the linear 2D PCA manifold (PCA2D).
proposed generalization of PCA [20] based on a Weighted The “basal” breast cancer subtype is visualized
PCA increases robustness by assigning different weights to more adequately with ELMap2D and some
data objects based on their estimated relevancy. features of the distribution become better resolved
in comparison to PCA2D. Principal manifolds are
produced by the elastic maps algorithm. Data are
Software/source code available for public competition[18].
"ViSta: The Visual Statistics System"
(http://www.mdp.edu.ar/psicologia/vista/vista.htm) a free software that provides principal components
analysis, simple and multiple correspondence analysis.
"Spectramap" (http://www.coloritto.com) is software to create a biplot using principal components analysis,
correspondence analysis or spectral map analysis.
The Unscrambler is a multivariate analysis software enabling Principal Component Analysis (PCA) with PCA
Projection.
Computer Vision Library (http://sourceforge.net/projects/opencvlibrary/)
Multivariate Data Analysis Software (http://astro.u-strasbg.fr/~fmurtagh/mda-sw/)
In the MATLAB Statistics Toolbox, the functions princomp and wmspca give the principal
components, while the function pcares gives the residuals and reconstructed matrix for a low-rank PCA
approximation. Here is a link to a MATLAB implementation of PCA PcaPress
(http://www.utdallas.edu/~herve/abdi-PCA4Wiley.zip) .
NMath, a numerical library containing PCA for the .NET Framework.
in Octave, the free software equivalent to MATLAB, the function princomp
(http://octave.sourceforge.net/statistics/function/princomp.html) gives the principal component.
in the open source statistical package R, the functions princomp (http://stat.ethz.ch/R-manual/R-
patched/library/stats/html/princomp.html) and prcomp (http://stat.ethz.ch/R-manual/R-
patched/library/stats/html/prcomp.html) can be used for principal component analysis; prcomp uses
singular value decomposition which generally gives better numerical accuracy. Recently there has been an
explosion in implementations of principal component analysis in various R packages, generally in packages
for specific purposes. There is no meaning in giving a complete list here, see [1] (http://cran.r-
project.org/web/views/Multivariate.html) .
In XLMiner, the Principles Component tab can be used for principal component analysis.
In IDL, the principal components can be calculated using the function pcomp.
Weka computes principal components (javadoc
(http://weka.sourceforge.net/doc/weka/attributeSelection/PrincipalComponents.html) ).
Software for analyzing multivariate data with instant response using PCA (http://www.qlucore.com)
Notes
1. ^ Pearson, K. (1901). "On Lines and Planes of Closest Fit to Systems of Points in Space"
(https://docs.google.com/viewer?url=http%3A%2F%2Fstat.smmu.edu.cn%2Fhistory%2Fpearson1901.pdf) (PDF).
Philosophical Magazine 2 (6): 559–572. http://stat.smmu.edu.cn/history/pearson1901.pdf.
2. ^ Jolliffe I.T. Principal Component Analysis
(http://www.springer.com/west/home/new+%26+forthcoming+titles+%28default%29?SGWID=4-40356-22-
2285433-0) , Series: Springer Series in Statistics
(http://www.springer.com/west/home/statistics/statistical+theory+and+methods?SGWID=4-10129-69-173621571-
0) , 2nd ed., Springer, NY, 2002, XXIX, 487 p. 28 illus. ISBN 978-0-387-95442-4
3. ^ A. A. Miranda, Y. A. Le Borgne, and G. Bontempi. New Routes from Minimal Approximation Error to Principal
Components (https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.ulb.ac.be%2Fdi%2Fmap%2Fyleborgn%2Fpub%2FNPL_PCA_07.pdf) , Volume 27,
Number 3 / June, 2008, Neural Processing Letters, Springer
4. ^ Fukunaga, Keinosuke (1990). Introduction to Statistical Pattern Recognition (http://books.google.com/books?
visbn=0122698517) . Elsevier. http://books.google.com/books?visbn=0122698517.
5. ^ Jonathon Shlens, A Tutorial on Principal Component Analysis. (https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.snl.salk.edu%2F~shlens%2Fpca.pdf)
6. ^ A.A. Miranda, Y.-A. Le Borgne, and G. Bontempi. New Routes from Minimal Approximation Error to Principal
Components (https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.ulb.ac.be%2Fdi%2Fmap%2Fyleborgn%2Fpub%2FNPL_PCA_07.pdf) , Volume 27,
Number 3 / June, 2008, Neural Processing Letters, Springer
7. ^ eig function (http://www.mathworks.com/access/helpdesk/help/techdoc/ref/eig.html#998306) Matlab
documentation
8. ^ MATLAB PCA-based Face recognition software
(http://www.mathworks.com/matlabcentral/fileexchange/24634)
9. ^ Eigenvalues function (http://reference.wolfram.com/mathematica/ref/Eigenvalues.html) Mathematica
documentation
10. ^ a b Roweis, Sam. "EM Algorithms for PCA and SPCA." Advances in Neural Information Processing Systems.
Ed. Michael I. Jordan, Michael J. Kearns, and Sara A. Solla The MIT Press, 1998.
11. ^ Geladi, Paul; Kowalski, Bruce (1986). "Partial Least Squares Regression:A Tutorial". Analytica Chimica Acta
185: 1–17. doi:10.1016/0003-2670(86)80028-9 (http://dx.doi.org/10.1016%2F0003-2670%2886%2980028-9) .
12. ^ H. Zha, C. Ding, M. Gu, X. He and H.D. Simon. "Spectral Relaxation for K-means Clustering",
http://ranger.uta.edu/~chqding/papers/Zha-Kmeans.pdf, Neural Information Processing Systems vol.14 (NIPS
2001). pp. 1057–1064, Vancouver, Canada. Dec. 2001.
13. ^ C. Ding and X. He. "K-means Clustering via Principal Component Analysis". Proc. of Int'l Conf. Machine
Learning (ICML 2004), pp 225–232. July 2004. http://ranger.uta.edu/~chqding/papers/KmeansPCA1.pdf
14. ^ Benzécri, J.-P. (1973). L'Analyse des Données. Volume II. L'Analyse des Correspondances. Paris, France:
Dunod.
15. ^ Greenacre, Michael (1983). Theory and Applications of Correspondence Analysis. London: Academic Press.
ISBN 0-12-299050-1.
16. ^ Le Roux, Brigitte and Henry Rouanet (2004). Geometric Data Analysis, From Correspondence Analysis to
Structured Data Analysis. Dordrecht: Kluwer.
17. ^ A. N. Gorban, A. Y. Zinovyev, Principal Graphs and Manifolds (http://arxiv.org/abs/0809.0490) , In: Handbook
of Research on Machine Learning Applications and Trends: Algorithms, Methods and Techniques, Olivas E.S. et al
Eds. Information Science Reference, IGI Global: Hershey, PA, USA, 2009. 28-59.
18. ^ Wang, Y., Klijn, J.G., Zhang, Y., Sieuwerts, A.M., Look, M.P., Yang, F., Talantov, D., Timmermans, M.,
Meijer-van Gelder, M.E., Yu, J. et al.: Gene expression profiles to predict distant metastasis of lymph-node-
negative primary breast cancer. Lancet 365, 671-679 (2005); Data online
(http://www.ihes.fr/~zinovyev/princmanif2006/)
19. ^ A. Gorban, B. Kegl, D. Wunsch, A. Zinovyev (Eds.), Principal Manifolds for Data Visualisation and Dimension
Reduction, (http://pca.narod.ru/contentsgkwz.htm) LNCSE 58, Springer, Berlin – Heidelberg – New York, 2007.
ISBN 978-3-540-73749-0
20. ^ Kriegel, H. P.; Kröger, P.; Schubert, E.; Zimek, A. (2008). A General Framework for Increasing the Robustness
of PCA-Based Correlation Clustering Algorithms. 5069. pp. 418. doi:10.1007/978-3-540-69497-7_27
(http://dx.doi.org/10.1007%2F978-3-540-69497-7_27) .
See also
Sparse PCA
Biplot
Eigenface
Exploratory factor analysis (Wikiversity)
Factor analysis
Geometric data analysis
Factorial code
Independent component analysis
Time Adaptive Self-Organizing Map
Kernel PCA
Elastic map
Matrix decomposition
Nonlinear dimensionality reduction
Oja's rule
PCA network
PCA applied to yield curves
Point distribution model (PCA applied to morphometry and computer vision)
Principal component regression
Principal component analysis (Wikibooks)
Singular spectrum analysis
Singular value decomposition
Transform coding
Weighted least squares
Dynamic mode decomposition
References
Jolliffe, I. T. (1986). Principal Component Analysis
(http://www.springer.com/west/home/new+%26+forthcoming+titles+%28default%29?SGWID=4-40356-
22-2285433-0) . Springer-Verlag. pp. 487. doi:10.1007/b98835 (http://dx.doi.org/10.1007%2Fb98835) .
ISBN 978-0-387-95442-4.
http://www.springer.com/west/home/new+%26+forthcoming+titles+%28default%29?SGWID=4-40356-
22-2285433-0.
R. Kramer, Chemometric Techniques for Quantitative Analysis, (1998) Marcel–Dekker, ISBN 0-8247-
0198-4.
Shaw PJA, Multivariate statistics for the Environmental Sciences, (2003) Hodder-Arnold.
Patra sk et al., J Photochemistry & Photobiology A:Chemistry, (1999) 122:23–31
External links
The Most Representative Composite Rank Ordering of Multi-Attribute Objects by the Particle Swarm
Optimization (http://ideas.repec.org/p/pra/mprapa/12723.html)
Sub-Optimality of Rank Ordering of Objects on the Basis of the Leading Principal Component Factor
Scores (http://ssrn.com/abstract=1321369)
Spectroscopy and PCA (http://neon.otago.ac.nz/chemlect/chem306/pca/index.html)
An introduction and review of recent developments of PCA (https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.utdallas.edu%2F~herve%2Fabdi-wireCS-PCA2010-inpress.pdf)
An introductory explanation of PCA from StatSoft (http://www.statsoft.com/textbook/stfacan.html)
A Tutorial on Principal Component Analysis by Jonathon Shlens (https://docs.google.com/viewer?
url=http%3A%2F%2Fwww.snl.salk.edu%2F~shlens%2Fpca.pdf) (PDF)
Principal Component Analysis using Hebbian learning tutorial (http://blog.peltarion.com/2006/06/20/the-
talented-drhebb-part-2-pca/)
Presentation of Principal Component Analysis used in Biomedical Engineering (http://brandon-
merkl.blogspot.com/2006/04/principal-components-analysis.html)
Application to microarray and other biomedical data (http://public.lanl.gov/mewall/kluwer2002.html)
PCA in functional neuroimaging, free software (http://feinsteinneuroscience.org/)
Uncertainty estimation for PCA (http://www.chemometry.com/Research/PCA.html)
FactoMineR, an R package dedicated to exploratory multivariate analysis (http://factominer.free.fr/)
A web-site with presentations and open source software on exploratory multivariate data analysis
(http://www.datascope.be/)
EasyPCA, a very simple and small PCA program under the GPL license (http://transp-
or2.epfl.ch/pagesPerso/javierFiles/software.php)
R tutorial on cluster and principal component analysis including example data
(http://www.iiap.res.in/astrostat/)
Simple COV-based PCA using Eigen Template Library in C++
(http://codingplayground.blogspot.com/2010/01/pca-dimensional-reduction-in-eigen.html) by Antonio Gulli
www.powercam.cc/chli (http://www.powercam.cc/chli) by Cheng-Hsuan Li (A Chinese Tutorial on Kernel
Method, PCA, KPCA, LDA, GDA, and SVMs)
Non-technical introductions
A layman's introduction to principal component analysis (http://www.youtube.com/watch?

v=BfTMmoDFXyE) (video)
Step by step tutorial from KVL (http://www.models.kvl.dk/demo/pca/story/default.asp) (video
(http://www.youtube.com/watch?v=4pnQd6jnCWk) )
Retrieved from "http://en.wikipedia.org/wiki/Principal_component_analysis"
Categories: Multivariate statistics | Singular value decomposition | Data mining | Data analysis | Machine learning
This page was last modified on 11 September 2010 at 16:19.

Text is available under the Creative Commons Attribution-ShareAlike License; additional terms may apply.
See Terms of Use for details.
Wikipedia® is a registered trademark of the Wikimedia Foundation, Inc., a non-profit organization.

Principal component analysis

Uploaded by

Copyright:

Available Formats

Principal component analysis

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Principal component analysis

Uploaded by

Copyright:

Available Formats

9/13/2010 Principal component analysis - Wikipe…

Principal component analysis

Principal component analysis (PCA) involves a mathematical

PCA was invented in 1901 by Karl Pearson.[1] Now it is mostly used as

Table of symbols and abbreviations

matrix consisting of the set of all eigenvectors of C,

Properties and limitations of PCA

Assumption on the statistical importance of mean and covariance

Assumption that large variances have important dynamics

Computing PCA using the covariance method

Organize the data set

Write as column vectors, each of which has M rows.

Calculate the empirical mean

Find the empirical mean along each dimension m = 1, ..., M.

Calculate the deviations from the mean

where h is a 1 × N row vector of all 1s:

Find the covariance matrix

is the expected value operator,

Find the eigenvectors and eigenvalues of the covariance matrix

Compute the matrix V of eigenvectors which diagonalizes the covariance matrix C:

Matrix D will take the form of an M × M diagonal matrix, where

is the mth eigenvalue of the covariance matrix C, and

Rearrange the eigenvectors and eigenvalues

Compute the cumulative energy content for each eigenvector

Select a subset of the eigenvectors as basis vectors

Save the first L columns of V as the M × L matrix W:

Convert the source data to z-scores

Calculate the M × N z-score matrix:

Project the z-scores of the data onto the new basis

The projected vectors are the columns of the matrix

W* is the conjugate transpose of the eigenvector matrix.

Derivation of PCA using the covariance method

with the constraint that

By substitution, and matrix algebra, we obtain:

Rewrite P as d column vectors, so

Substituting into equation above, we obtain:

Notice that in , Pi is an eigenvector of the covariance matrix of X. Therefore, by finding the

Computing principal components with expectation maximization

for each row

The NIPALS method

Relation between PCA and K-means clustering

Most of the modern methods for nonlinear dimensionality

Robustness - Weighted PCA Linear PCA versus nonlinear Principal

"ViSta: The Visual Statistics System"

A layman's introduction to principal component analysis (http://www.youtube.com/watch?

This page was last modified on 11 September 2010 at 16:19.

You might also like