Principal component analysis

From Wikipedia, the free encyclopedia

(Redirected from Principal Component Analysis)
Jump to: navigation, search
"KLT" redirects here. For the Kanade–Lucas–Tomasi feature tracker used in computer vision,
see Kanade–Lucas–Tomasi Feature Tracker.
This article or section appears to contradict itself. Please see its talk page for more
information. (May 2009)

PCA of a multivariate Gaussian distribution centered at (1,3) with a standard deviation of 3 in

roughly the (0.878, 0.478) direction and of 1 in the orthogonal direction.

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 involves the calculation of the eigenvalue
decomposition of a data covariance matrix or singular value decomposition of a data matrix,
usually after mean centering the data for each attribute. The results of a PCA are usually
discussed in terms of component scores and loadings (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
[hide]

• 1 Details
• 2 Discussion
• 3 Table of symbols and abbreviations
• 4 Properties and limitations of PCA
• 5 Computing PCA using the covariance method
o 5.1 Organize the data set
o 5.2 Calculate the empirical mean
o 5.3 Calculate the deviations from the mean
o 5.4 Find the covariance matrix
o 5.5 Find the eigenvectors and eigenvalues of the covariance matrix
o 5.6 Rearrange the eigenvectors and eigenvalues
o 5.7 Compute the cumulative energy content for each eigenvector
o 5.8 Select a subset of the eigenvectors as basis vectors
o 5.9 Convert the source data to z-scores
o 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
o 7.1 The NIPALS method
• 8 Relation between PCA and K-means clustering
• 9 Correspondence analysis
• 10 Generalizations
o 10.1 Nonlinear generalizations
o 10.2 Higher order
o 10.3 Robustness - Weighted PCA
• 11 Software/source code
• 12 Notes
• 13 See also
• 14 References
• 15 External links

o 15.1 Non-technical introductions

[edit] 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.

[edit] 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.

[edit] 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
scalar
vector (dimension)
the number of dimensions in the
dimensionally reduced subspace, scalar

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.

[edit] 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.

[edit] Computing PCA using the covariance method

Following is a detailed description of PCA using the covariance method (see also here). 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:

[edit] 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.

[edit] 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.

[edit] 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:

[edit] 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.

[edit] 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.

[edit] 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.

[edit] 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:

[edit] 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

[edit] 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!

[edit] 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.

[edit] 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.

[edit] 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],

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).

[edit] 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.

[edit] 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.

[edit] 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].

[edit] Generalizations
[edit] Nonlinear generalizations

Linear PCA versus nonlinear Principal Manifolds[17] for visualization of breast cancer microarray
data: a) Configuration of nodes and 2D Principal Surface in the 3D PCA linear manifold. The
dataset is curved and can not be mapped adequately on a 2D principal plane; b) The distribution
in the internal 2D non-linear principal surface coordinates (ELMap2D) together with an
estimation of the density of points; c) The same as b), but for the linear 2D PCA manifold
(PCA2D). The “basal” breast cancer subtype is visualized more adequately with ELMap2D and
some features of the distribution become better resolved in comparison to PCA2D. Principal
manifolds are produced by the elastic maps algorithm. Data are available for public
competition[18].

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.

[edit] 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.

[edit] Robustness - Weighted PCA

While PCA finds the mathematically optimal method (as in minimizing the squared error), it is
sensitive to outliers in the data that produce large errors PCA tries to avoid. It therefore is
common practise to remove outliers before computing PCA. However, in some contexts, outliers
can be difficult to identify. For example in data mining algorithms like correlation clustering, the
assignment of points to clusters and outliers is not known beforehand. A recently proposed
generalization of PCA [20] based on a Weighted PCA increases robustness by assigning different
weights to data objects based on their estimated relevancy.