Principal component analysis

Last updated
PCA of a multivariate Gaussian distribution centered at (1,3) with a standard deviation of 3 in roughly the (0.866, 0.5) direction and of 1 in the orthogonal direction. The vectors shown are the eigenvectors of the covariance matrix scaled by the square root of the corresponding eigenvalue, and shifted so their tails are at the mean. GaussianScatterPCA.svg
PCA of a multivariate Gaussian distribution centered at (1,3) with a standard deviation of 3 in roughly the (0.866, 0.5) direction and of 1 in the orthogonal direction. The vectors shown are the eigenvectors of the covariance matrix scaled by the square root of the corresponding eigenvalue, and shifted so their tails are at the mean.

The principal components of a collection of points in a real coordinate space are a sequence of unit vectors, where the -th vector is the direction of a line that best fits the data while being orthogonal to the first vectors. Here, a best-fitting line is defined as one that minimizes the average squared distance from the points to the line. These directions constitute an orthonormal basis in which different individual dimensions of the data are linearly uncorrelated. Principal component analysis (PCA) is the process of computing the principal components and using them to perform a change of basis on the data, sometimes using only the first few principal components and ignoring the rest.


In data analysis, the first principal component of a set of variables, presumed to be jointly normally distributed, is the derived variable formed as a linear combination of the original variables that explains the most variance. The second principal component explains the most variance in what is left once the effect of the first component is removed, and we may proceed through iterations until all the variance is explained. PCA is most commonly used when many of the variables are highly correlated with each other and it is desirable to reduce their number to an independent set.

PCA is used in exploratory data analysis and for making predictive models. It is commonly used for dimensionality reduction by projecting each data point onto only the first few principal components to obtain lower-dimensional data while preserving as much of the data's variation as possible. The first principal component can equivalently be defined as a direction that maximizes the variance of the projected data. The -th principal component can be taken as a direction orthogonal to the first principal components that maximizes the variance of the projected data.

For either objective, it can be shown that the principal components are eigenvectors of the data's covariance matrix. Thus, the principal components are often computed by eigendecomposition of the data covariance matrix or singular value decomposition of the data matrix. PCA is the simplest of the true eigenvector-based multivariate analyses and is closely related to factor analysis. Factor analysis typically incorporates more domain specific assumptions about the underlying structure and solves eigenvectors of a slightly different matrix. PCA is also related to canonical correlation analysis (CCA). CCA defines coordinate systems that optimally describe the cross-covariance between two datasets while PCA defines a new orthogonal coordinate system that optimally describes variance in a single dataset. [1] [2] [3] [4] Robust and L1-norm-based variants of standard PCA have also been proposed. [5] [6] [4]


PCA was invented in 1901 by Karl Pearson, [7] as an analogue of the principal axis theorem in mechanics; it was later independently developed and named by Harold Hotelling in the 1930s. [8] Depending on the field of application, it is also named the discrete Karhunen–Loève transform (KLT) in signal processing, the Hotelling transform in multivariate quality control, proper orthogonal decomposition (POD) in mechanical engineering, singular value decomposition (SVD) of X (invented in the last quarter of the 19th century [9] ), eigenvalue decomposition (EVD) of XTX in linear algebra, factor analysis (for a discussion of the differences between PCA and factor analysis see Ch. 7 of Jolliffe's Principal Component Analysis), [10] Eckart–Young theorem (Harman, 1960), or empirical orthogonal functions (EOF) in meteorological science, empirical eigenfunction decomposition (Sirovich, 1987), empirical component analysis (Lorenz, 1956), quasiharmonic modes (Brooks et al., 1988), spectral decomposition in noise and vibration, and empirical modal analysis in structural dynamics.


PCA can be thought of as fitting a p-dimensional ellipsoid to the data, where each axis of the ellipsoid represents a principal component. If some axis of the ellipsoid is small, then the variance along that axis is also small.

To find the axes of the ellipsoid, we must first center the values of each variable in the dataset on 0 by subtracting the mean of the variable's observed values from each of those values. These transformed values are used instead of the original observed values for each of the variables. Then, we compute the covariance matrix of the data and calculate the eigenvalues and corresponding eigenvectors of this covariance matrix. Then we must normalize each of the orthogonal eigenvectors to turn them into unit vectors. Once this is done, each of the mutually orthogonal unit eigenvectors can be interpreted as an axis of the ellipsoid fitted to the data. This choice of basis will transform our covariance matrix into a diagonalised form with the diagonal elements representing the variance of each axis. The proportion of the variance that each eigenvector represents can be calculated by dividing the eigenvalue corresponding to that eigenvector by the sum of all eigenvalues.

Biplots and scree plots (degree of explained variance) are used to explain findings of the PCA.

The above picture is of a scree plot that is meant to help interpret the PCA and decide how many components to retain. The start of the bend in the line (point of inflexion) should indicate how many components are retained, hence in this example, three factors should be retained. SCREE plot.jpg
The above picture is of a scree plot that is meant to help interpret the PCA and decide how many components to retain. The start of the bend in the line (point of inflexion) should indicate how many components are retained, hence in this example, three factors should be retained.


PCA is defined as an orthogonal linear transformation that transforms the data to a new coordinate system such that the greatest variance by some scalar 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. [10] [ page needed ]

Consider an data matrix, X, with column-wise zero empirical mean (the sample mean of each column has been shifted to zero), where each of the n rows represents a different repetition of the experiment, and each of the p columns gives a particular kind of feature (say, the results from a particular sensor).

Mathematically, the transformation is defined by a set of size of p-dimensional vectors of weights or coefficients that map each row vector of X to a new vector of principal component scores, given by

in such a way that the individual variables of t considered over the data set successively inherit the maximum possible variance from X, with each coefficient vector w constrained to be a unit vector (where is usually selected to be strictly less than to reduce dimensionality).

First component

In order to maximize variance, the first weight vector w(1) thus has to satisfy

Equivalently, writing this in matrix form gives

Since w(1) has been defined to be a unit vector, it equivalently also satisfies

The quantity to be maximised can be recognised as a Rayleigh quotient. A standard result for a positive semidefinite matrix such as XTX is that the quotient's maximum possible value is the largest eigenvalue of the matrix, which occurs when w is the corresponding eigenvector.

With w(1) found, the first principal component of a data vector x(i) can then be given as a score t1(i) = x(i)w(1) in the transformed co-ordinates, or as the corresponding vector in the original variables, {x(i)w(1)} w(1).

Further components

The k-th component can be found by subtracting the first k  1 principal components from X:

and then finding the weight vector which extracts the maximum variance from this new data matrix

It turns out that this gives the remaining eigenvectors of XTX, with the maximum values for the quantity in brackets given by their corresponding eigenvalues. Thus the weight vectors are eigenvectors of XTX.

The k-th principal component of a data vector x(i) can therefore be given as a score tk(i) = x(i)w(k) in the transformed coordinates, or as the corresponding vector in the space of the original variables, {x(i)w(k)} w(k), where w(k) is the kth eigenvector of XTX.

The full principal components decomposition of X can therefore be given as

where W is a p-by-p matrix of weights whose columns are the eigenvectors of XTX. The transpose of W is sometimes called the whitening or sphering transformation. Columns of W multiplied by the square root of corresponding eigenvalues, that is, eigenvectors scaled up by the variances, are called loadings in PCA or in Factor analysis.


XTX itself can be recognized as proportional to the empirical sample covariance matrix of the dataset XT. [10] :30–31

The sample covariance Q between two of the different principal components over the dataset is given by:

where the eigenvalue property of w(k) has been used to move from line 2 to line 3. However eigenvectors w(j) and w(k) corresponding to eigenvalues of a symmetric matrix are orthogonal (if the eigenvalues are different), or can be orthogonalised (if the vectors happen to share an equal repeated value). The product in the final line is therefore zero; there is no sample covariance between different principal components over the dataset.

Another way to characterise the principal components transformation is therefore as the transformation to coordinates which diagonalise the empirical sample covariance matrix.

In matrix form, the empirical covariance matrix for the original variables can be written

The empirical covariance matrix between the principal components becomes

where Λ is the diagonal matrix of eigenvalues λ(k) of XTX. λ(k) is equal to the sum of the squares over the dataset associated with each component k, that is, λ(k) = Σitk2(i) = Σi (x(i)w(k))2.

Dimensionality reduction

The transformation T = XW maps a data vector x(i) from an original space of p variables to a new space of p variables which are uncorrelated over the dataset. However, not all the principal components need to be kept. Keeping only the first L principal components, produced by using only the first L eigenvectors, gives the truncated transformation

where the matrix TL now has n rows but only L columns. In other words, PCA learns a linear transformation where the columns of p × L matrix form an orthogonal basis for the L features (the components of representation t) that are decorrelated. [11] By construction, of all the transformed data matrices with only L columns, this score matrix maximises the variance in the original data that has been preserved, while minimising the total squared reconstruction error or .

A principal components analysis scatterplot of Y-STR haplotypes calculated from repeat-count values for 37 Y-chromosomal STR markers from 354 individuals.
PCA has successfully found linear combinations of the markers that separate out different clusters corresponding to different lines of individuals' Y-chromosomal genetic descent. PCA of Haplogroup J using 37 STRs.png
A principal components analysis scatterplot of Y-STR haplotypes calculated from repeat-count values for 37 Y-chromosomal STR markers from 354 individuals.
PCA has successfully found linear combinations of the markers that separate out different clusters corresponding to different lines of individuals' Y-chromosomal genetic descent.

Such dimensionality reduction can be a very useful step for visualising and processing high-dimensional datasets, while still retaining as much of the variance in the dataset as possible. For example, selecting L = 2 and keeping only the first two principal components finds the two-dimensional plane through the high-dimensional dataset in which the data is most spread out, so if the data contains clusters these too may be most spread out, and therefore most visible to be plotted out in a two-dimensional diagram; whereas if two directions through the data (or two of the original variables) are chosen at random, the clusters may be much less spread apart from each other, and may in fact be much more likely to substantially overlay each other, making them indistinguishable.

Similarly, in regression analysis, the larger the number of explanatory variables allowed, the greater is the chance of overfitting the model, producing conclusions that fail to generalise to other datasets. One approach, especially when there are strong correlations between different possible explanatory variables, is to reduce them to a few principal components and then run the regression against them, a method called principal component regression.

Dimensionality reduction may also be appropriate when the variables in a dataset are noisy. If each column of the dataset contains independent identically distributed Gaussian noise, then the columns of T will also contain similarly identically distributed Gaussian noise (such a distribution is invariant under the effects of the matrix W, which can be thought of as a high-dimensional rotation of the co-ordinate axes). However, with more of the total variance concentrated in the first few principal components compared to the same noise variance, the proportionate effect of the noise is less—the first few components achieve a higher signal-to-noise ratio. PCA thus can have the effect of concentrating much of the signal into the first few principal components, which can usefully be captured by dimensionality reduction; while the later principal components may be dominated by noise, and so disposed of without great loss. If the dataset is not too large, the significance of the principal components can be tested using parametric bootstrap, as an aid in determining how many principal components to retain. [12]

Singular value decomposition

The principal components transformation can also be associated with another matrix factorization, the singular value decomposition (SVD) of X,

Here Σ is an n-by-p rectangular diagonal matrix of positive numbers σ(k), called the singular values of X; U is an n-by-n matrix, the columns of which are orthogonal unit vectors of length n called the left singular vectors of X; and W is a p-by-p whose columns are orthogonal unit vectors of length p and called the right singular vectors of X.

In terms of this factorization, the matrix XTX can be written

where is the square diagonal matrix with the singular values of X and the excess zeros chopped off that satisfies. Comparison with the eigenvector factorization of XTX establishes that the right singular vectors W of X are equivalent to the eigenvectors of XTX, while the singular values σ(k) of are equal to the square-root of the eigenvalues λ(k) of XTX.

Using the singular value decomposition the score matrix T can be written

so each column of T is given by one of the left singular vectors of X multiplied by the corresponding singular value. This form is also the polar decomposition of T.

Efficient algorithms exist to calculate the SVD of X without having to form the matrix XTX, so computing the SVD is now the standard way to calculate a principal components analysis from a data matrix[ citation needed ], unless only a handful of components are required.

As with the eigen-decomposition, a truncated n × L score matrix TL can be obtained by considering only the first L largest singular values and their singular vectors:

The truncation of a matrix M or T using a truncated singular value decomposition in this way produces a truncated matrix that is the nearest possible matrix of rank L to the original matrix, in the sense of the difference between the two having the smallest possible Frobenius norm, a result known as the Eckart–Young theorem [1936].

Further considerations

Given a set of points in Euclidean space, the first principal component corresponds to a line that passes through the multidimensional mean and minimizes the sum of squares of the distances of the points from the line [ disputed ]. The second principal component corresponds to the same concept after all correlation with the first principal component has been subtracted from the points. The singular values (in Σ) are the square roots of the eigenvalues of the matrix XTX. Each eigenvalue is proportional to the portion of the "variance" (more correctly of the sum of the squared distances of the points from their multidimensional mean) that is associated with each eigenvector. The sum of all the eigenvalues is equal to the sum of the squared distances of the points from their multidimensional mean. PCA essentially rotates the set of points around their mean in order to align with the principal components. This moves as much of the variance as possible (using an orthogonal transformation) into the first few dimensions. The values in the remaining dimensions, therefore, tend to be small and may be dropped with minimal loss of information (see below). PCA is often used in this manner for dimensionality reduction. PCA has the distinction of being the optimal orthogonal transformation for keeping the subspace that has largest "variance" (as defined above). This advantage, however, comes at the price of greater computational requirements if compared, for example, and when applicable, to the discrete cosine transform, and in particular to the DCT-II which is simply known as the "DCT". Nonlinear dimensionality reduction techniques tend to be more computationally demanding than PCA.

PCA is sensitive to the scaling of the variables. If we have just two variables and they have the same sample variance and are completely correlated, then the PCA will entail a rotation by 45° and the "weights" (they are the cosines of rotation) for the two variables with respect to the principal component will be equal. But if we multiply all values of the first variable by 100, then the first principal component will be almost the same as that variable, with a small contribution from the other variable, whereas the second component will be almost aligned with the second original variable. This means that whenever the different variables have different units (like temperature and mass), PCA is a somewhat arbitrary method of analysis. (Different results would be obtained if one used Fahrenheit rather than Celsius for example.) Pearson's original paper was entitled "On Lines and Planes of Closest Fit to Systems of Points in Space" – "in space" implies physical Euclidean space where such concerns do not arise. One way of making the PCA less arbitrary is to use variables scaled so as to have unit variance, by standardizing the data and hence use the autocorrelation matrix instead of the autocovariance matrix as a basis for PCA. However, this compresses (or expands) the fluctuations in all dimensions of the signal space to unit variance.

Mean subtraction (a.k.a. "mean centering") is necessary for performing classical PCA to ensure that the first principal component describes the direction of maximum variance. If mean subtraction is not performed, the first principal component might instead correspond more or less 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. [13]

Mean-centering is unnecessary if performing a principal components analysis on a correlation matrix, as the data are already centered after calculating correlations. Correlations are derived from the cross-product of two standard scores (Z-scores) or statistical moments (hence the name: Pearson Product-Moment Correlation). Also see the article by Kromrey & Foster-Johnson (1998) on "Mean-centering in Moderated Regression: Much Ado About Nothing".

PCA is a popular primary technique in pattern recognition. It is not, however, optimized for class separability. [14] However, it has been used to quantify the distance between two or more classes by calculating center of mass for each class in principal component space and reporting Euclidean distance between center of mass of two or more classes. [15] The linear discriminant analysis is an alternative which is optimized for class separability.

Table of symbols and abbreviations

data matrix, consisting of the set of all data vectors, one vector per row
the number of row vectors in the data setscalar
the number of elements in each row vector (dimension)scalar
the number of dimensions in the dimensionally reduced subspace, scalar
vector of empirical means, one mean for each column j of the data matrix
vector of empirical standard deviations, one standard deviation for each column j of the data matrix
vector of all 1's
deviations from the mean of each column j 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 ( note used above )
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 row 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


Some properties of PCA include: [10] [ page needed ]

Property 1: For any integer q, 1 ≤ qp, consider the orthogonal linear transformation
where is a q-element vector and is a (q × p) matrix, and let be the variance-covariance matrix for . Then the trace of , denoted , is maximized by taking , where consists of the first q columns of is the transposition of .
Property 2: Consider again the orthonormal transformation
with and defined as before. Then is minimized by taking where consists of the last q columns of .

The statistical implication of this property is that the last few PCs are not simply unstructured left-overs after removing the important PCs. Because these last PCs have variances as small as possible they are useful in their own right. They can help to detect unsuspected near-constant linear relationships between the elements of x, and they may also be useful in regression, in selecting a subset of variables from x, and in outlier detection.

Property 3: (Spectral decomposition of Σ)

Before we look at its usage, we first look at diagonal elements,

Then, perhaps the main statistical implication of the result is that not only can we decompose the combined variances of all the elements of x into decreasing contributions due to each PC, but we can also decompose the whole covariance matrix into contributions from each PC. Although not strictly decreasing, the elements of will tend to become smaller as increases, as is nonincreasing for increasing , whereas the elements of tend to stay about the same size because of the normalization constraints: .


As noted above, the results of PCA depend on the scaling of the variables. This can be cured by scaling each feature by its standard deviation, so that one ends up with dimensionless features with unital variance. [16]

The applicability of PCA as described above is limited by certain (tacit) assumptions [17] made in its derivation. In particular, PCA can capture linear correlations between the features but fails when this assumption is violated (see Figure 6a in the reference). In some cases, coordinate transformations can restore the linearity assumption and PCA can then be applied (see kernel PCA).

Another limitation is the mean-removal process before constructing the covariance matrix for PCA. In fields such as astronomy, all the signals are non-negative, and the mean-removal process will force the mean of some astrophysical exposures to be zero, which consequently creates unphysical negative fluxes, [18] and forward modeling has to be performed to recover the true magnitude of the signals. [19] As an alternative method, non-negative matrix factorization focusing only on the non-negative elements in the matrices, which is well-suited for astrophysical observations. [20] [21] [22] See more at Relation between PCA and Non-negative Matrix Factorization.

PCA is at a disadvantage if the data has not been standardized before applying the algorithm to it. PCA transforms original data into data that is relevant to the principal components of that data, which means that the new data variables cannot be interpreted in the same ways that the originals were. They are linear interpretations of the original variables. Also, if PCA is not performed properly, there is a high likelihood of information loss. [23]

PCA relies on a linear model. If a dataset has a pattern hidden inside it that is nonlinear, then PCA can actually steer the analysis in the complete opposite direction of progress. [24] [ page needed ] Researchers at Kansas State University discovered that the sampling error in their experiments impacted the bias of PCA results. "If the number of subjects or blocks is smaller than 30, and/or the researcher is interested in PC's beyond the first, it may be better to first correct for the serial correlation, before PCA is conducted". [25] The researchers at Kansas State also found that PCA could be "seriously biased if the autocorrelation structure of the data is not correctly handled". [25]

PCA and information theory

Dimensionality reduction results in a loss of information, in general. PCA-based dimensionality reduction tends to minimize that information loss, under certain signal and noise models.

Under the assumption that

that is, that the data vector is the sum of the desired information-bearing signal and a noise signal one can show that PCA can be optimal for dimensionality reduction, from an information-theoretic point-of-view.

In particular, Linsker showed that if is Gaussian and is Gaussian noise with a covariance matrix proportional to the identity matrix, the PCA maximizes the mutual information between the desired information and the dimensionality-reduced output . [26]

If the noise is still Gaussian and has a covariance matrix proportional to the identity matrix (that is, the components of the vector are iid), but the information-bearing signal is non-Gaussian (which is a common scenario), PCA at least minimizes an upper bound on the information loss, which is defined as [27] [28]

The optimality of PCA is also preserved if the noise is iid and at least more Gaussian (in terms of the Kullback–Leibler divergence) than the information-bearing signal . [29] In general, even if the above signal model holds, PCA loses its information-theoretic optimality as soon as the noise becomes dependent.

Computing PCA using the covariance method

The following is a detailed description of PCA using the covariance method (see also here) as opposed to the correlation method. [30]

The goal is to transform a given data set X of dimension p 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 p variables, and you want to reduce the data so that each observation can be described with only L variables, L < p. Suppose further, that the data are arranged as a set of n data vectors with each representing a single grouped observation of the p variables.

Calculate the empirical mean
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. [31] Hence we proceed by centering the data as follows:

In some applications, each variable (column of B) may also be scaled to have a variance equal to 1 (see Z-score). [32] This step affects the calculated principal components, but makes them independent of the units used to measure the different variables.

Find the covariance matrix
Find the eigenvectors and eigenvalues of the covariance matrix
Rearrange the eigenvectors and eigenvalues
Compute the cumulative energy content for each eigenvector
Select a subset of the eigenvectors as basis vectors
Project the data onto the new basis

That is, the first column of is the projection of the data points onto the first principal component, the second column is the projection onto the second principal component, etc.

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 d × d orthonormal transformation matrix P so that PX has a diagonal covariance matrix (that is, PX is a random vector with all its distinct components pairwise uncorrelated).

A quick computation assuming were unitary yields:

Hence holds if and only if were diagonalisable by .

This is very constructive, as cov(X) is guaranteed to be a non-negative definite matrix and thus is guaranteed to be diagonalisable by some unitary matrix.

Covariance-free computation

In practical implementations, especially with high dimensional data (large p), the naive covariance method is rarely used because it is not efficient due to high computational and memory costs of explicitly determining the covariance matrix. The covariance-free approach avoids the np2 operations of explicitly calculating and storing the covariance matrix XTX, instead utilizing one of matrix-free methods, for example, based on the function evaluating the product XT(X r) at the cost of 2np operations.

Iterative computation

One way to compute the first principal component efficiently [37] is shown in the following pseudo-code, for a data matrix X with zero mean, without ever computing its covariance matrix.

r = a random vector of length pr = r / norm(r) do c times:       s = 0 (a vector of length p)       for each row x in Xs = s + (xr) xλ = rTs// λ is the eigenvalueerror = |λ ⋅ rs|r = s / norm(s)exit if error < tolerance return λ, r

This power iteration algorithm simply calculates the vector XT(X r), normalizes, and places the result back in r. The eigenvalue is approximated by rT (XTX) r, which is the Rayleigh quotient on the unit vector r for the covariance matrix XTX . If the largest singular value is well separated from the next largest one, the vector r gets close to the first principal component of X within the number of iterations c, which is small relative to p, at the total cost 2cnp. The power iteration convergence can be accelerated without noticeably sacrificing the small cost per iteration using more advanced matrix-free methods, such as the Lanczos algorithm or the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method.

Subsequent principal components can be computed one-by-one via deflation or simultaneously as a block. In the former approach, imprecisions in already computed approximate principal components additively affect the accuracy of the subsequently computed principal components, thus increasing the error with every new computation. The latter approach in the block power method replaces single-vectors r and s with block-vectors, matrices R and S. Every column of R approximates one of the leading principal components, while all columns are iterated simultaneously. The main calculation is evaluation of the product XT(X R). Implemented, for example, in LOBPCG, efficient blocking eliminates the accumulation of the errors, allows using high-level BLAS matrix-matrix product functions, and typically leads to faster convergence, compared to the single-vector one-by-one technique.

The NIPALS method

Non-linear iterative partial least squares (NIPALS) is a variant the classical power iteration with matrix deflation by subtraction implemented for computing the first few components in a principal component or partial least squares analysis. For very-high-dimensional datasets, such as those generated in the *omics sciences (for example, genomics, metabolomics) it is usually only necessary to compute the first few PCs. The non-linear iterative partial least squares (NIPALS) algorithm updates iterative approximations to the leading scores and loadings t1 and r1T by the power iteration multiplying on every iteration by X on the left and on the right, that is, calculation of the covariance matrix is avoided, just as in the matrix-free implementation of the power iterations to XTX, based on the function evaluating the product XT(X r) = ((X r)TX)T.

The matrix deflation by subtraction is performed by subtracting the outer product, t1r1T from X leaving the deflated residual matrix used to calculate the subsequent leading PCs. [38] For large data matrices, or matrices that have a high degree of column collinearity, NIPALS suffers from loss of orthogonality of PCs due to machine precision round-off errors accumulated in each iteration and matrix deflation by subtraction. [39] A Gram–Schmidt re-orthogonalization algorithm is applied to both the scores and the loadings at each iteration step to eliminate this loss of orthogonality. [40] NIPALS reliance on single-vector multiplications cannot take advantage of high-level BLAS and results in slow convergence for clustered leading singular values—both these deficiencies are resolved in more sophisticated matrix-free block solvers, such as the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method.

Online/sequential estimation

In an "online" or "streaming" situation with data arriving piece by piece rather than being stored in a single batch, it is useful to make an estimate of the PCA projection that can be updated sequentially. This can be done efficiently, but requires different algorithms. [41]

PCA and qualitative variables

In PCA, it is common that we want to introduce qualitative variables as supplementary elements. For example, many quantitative variables have been measured on plants. For these plants, some qualitative variables are available as, for example, the species to which the plant belongs. These data were subjected to PCA for quantitative variables. When analyzing the results, it is natural to connect the principal components to the qualitative variable species. For this, the following results are produced.

These results are what is called introducing a qualitative variable as supplementary element. This procedure is detailed in and Husson, Lê & Pagès 2009 and Pagès 2013. Few software offer this option in an "automatic" way. This is the case of SPAD that historically, following the work of Ludovic Lebart, was the first to propose this option, and the R package FactoMineR.



The earliest application of factor analysis was in locating and measuring components of human intelligence. it was believed that intelligence had various uncorrelated components such as spatial intelligence, verbal intelligence, induction, deduction etc and that scores on these could be adduced by factor analysis from results on various tests, to give a single index known as the Intelligence Quotient (IQ). The pioneering statistical psychologist Spearman actually developed factor analysis in 1904 for his two-factor theory of intelligence, adding a formal technique to the science of psychometrics. In 1924 Thurstone looked for 56 factors of intelligence, developing the notion of Mental Age. Standard IQ tests today are based on this early work. [42]

Residential differentiation

In 1949, Shevky and Williams introduced the theory of factorial ecology, which dominated studies of residential differentiation from the 1950s to the 1970s. [43] Neighbourhoods in a city were recognizable or could be distinguished from one another by various characteristics which could be reduced to three by factor analysis. These were known as 'social rank' (an index of occupational status), 'familism' or family size, and 'ethnicity; Cluster analysis could then be applied to divide the city into clusters or precincts according to values of the three key factor variables. An extensive literature developed around factorial ecology in urban geography, but the approach went out of fashion after 1980 as being methodologically primitive and having little place in postmodern geographical paradigms.

One of the problems with factor analysis has always been finding convincing names for the various artificial factors. In 2000, Flood revived the factorial ecology approach to show that principal components analysis actually gave meaningful answers directly, without resorting to factor rotation. The principal components were actually dual variables or shadow prices of 'forces' pushing people together or apart in cities. The first component was 'accessibility', the classic trade-off between demand for travel and demand for space, around which classical urban economics is based. The next two components were 'disadvantage', which keeps people of similar status in separate neighbourhoods (mediated by planning), and ethnicity, where people of similar ethnic backgrounds try to co-locate. [44]

About the same time, the Australian Bureau of Statistics defined distinct indexes of advantage and disadvantage taking the first principal component of sets of key variables that were thought to be important. These SEIFA indexes are regularly published for various jurisdictions, and are used frequently in spatial analysis. [45]

Development indexes

PCA has been the only formal method available for the development of indexes, which are otherwise a hit-or-miss ad hoc undertaking.

The City Development Index was developed by PCA from about 200 indicators of city outcomes in a 1996 survey of 254 global cities. The first principal component was subject to iterative regression, adding the original variables singly until about 90% of its variation was accounted for. The index ultimately used about 15 indicators but was a good predictor of many more variables. Its comparative value agreed very well with a subjective assessment of the condition of each city. The coefficients on items of infrastructure were roughly proportional to the average costs of providing the underlying services, suggesting the Index was actually a measure of effective physical and social investment in the city.

The country-level Human Development Index (HDI) from UNDP, which has been published since 1990 and is very extensively used in development studies, [46] has very similar coefficients on similar indicators, strongly suggesting it was originally constructed using PCA.

Population genetics

In 1978 Cavalli-Sforza and others pioneered the use of principal components analysis (PCA) to summarise data on variation in human gene frequencies across regions. The components showed distinctive patterns, including gradients and sinusoidal waves. They interpreted these patterns as resulting from specific ancient migration events.

Since then, PCA has been ubiquitous in population genetics, with thousands of papers using PCA as a display mechanism. Genetics varies largely according to proximity, so the first two principal components actually show spatial distribution and may be used to map the relative geographical location of different population groups, thereby showing individuals who have wandered from their original locations. [47]

PCA in genetics has been technically controversial, in that the technique has been performed on discrete non-normal variables and often on binary allele markers. The lack of any measures of standard error in PCA are also an impediment to more consistent usage.

Market research and indexes of attitude

Market research has been an extensive user of PCA. It is used to develop customer satisfaction or customer loyalty scores for products, and with clustering, to develop market segments that may be targeted with advertising campaigns, in much the same way as factorial ecology will locate geographical areas with similar characteristics. [48]

PCA rapidly transforms large amounts of data into smaller, easier-to-digest variables that can be more rapidly and readily analyzed. In any consumer questionnaire, there are series of questions designed to elicit consumer attitudes, and principal components seek out latent variables underlying these attitudes. For example, the Oxford Internet Survey in 2013 asked 2000 people about their attitudes and beliefs, and from these analysts extracted four principal component dimensions, which they identified as 'escape', 'social networking', 'efficiency', and 'problem creating'. [49]

Another example from Joe Flood in 2008 extracted an attitudinal index toward housing from 28 attitude questions in a national survey of 2697 households in Australia. The first principal component represented a general attitude toward property and home ownership. The index, or the attitude questions it embodied, could be fed into a General Linear Model of tenure choice. The strongest determinant of private renting by far was the attitude index, rather than income, marital status or household type. [50]

Quantitative finance

In quantitative finance, principal component analysis can be directly applied to the risk management of interest rate derivative portfolios. [51] Trading multiple swap instruments which are usually a function of 30–500 other market quotable swap instruments is sought to be reduced to usually 3 or 4 principal components, representing the path of interest rates on a macro basis. Converting risks to be represented as those to factor loadings (or multipliers) provides assessments and understanding beyond that available to simply collectively viewing risks to individual 30–500 buckets.

PCA has also been applied to equity portfolios in a similar fashion, [52] both to portfolio risk and to risk return. One application is to reduce portfolio risk, where allocation strategies are applied to the "principal portfolios" instead of the underlying stocks. [53] A second is to enhance portfolio return, using the principal components to select stocks with upside potential.[ citation needed ]


A variant of principal components analysis is used in neuroscience to identify the specific properties of a stimulus that increases a neuron's probability of generating an action potential. [54] This technique is known as spike-triggered covariance analysis. In a typical application an experimenter presents a white noise process as a stimulus (usually either as a sensory input to a test subject, or as a current injected directly into the neuron) and records a train of action potentials, or spikes, produced by the neuron as a result. Presumably, certain features of the stimulus make the neuron more likely to spike. In order to extract these features, the experimenter calculates the covariance matrix of the spike-triggered ensemble, the set of all stimuli (defined and discretized over a finite time window, typically on the order of 100 ms) that immediately preceded a spike. The eigenvectors of the difference between the spike-triggered covariance matrix and the covariance matrix of the prior stimulus ensemble (the set of all stimuli, defined over the same length time window) then indicate the directions in the space of stimuli along which the variance of the spike-triggered ensemble differed the most from that of the prior stimulus ensemble. Specifically, the eigenvectors with the largest positive eigenvalues correspond to the directions along which the variance of the spike-triggered ensemble showed the largest positive change compared to the varince of the prior. Since these were the directions in which varying the stimulus led to a spike, they are often good approximations of the sought after relevant stimulus features.

In neuroscience, PCA is also used to discern the identity of a neuron from the shape of its action potential. Spike sorting is an important procedure because extracellular recording techniques often pick up signals from more than one neuron. In spike sorting, one first uses PCA to reduce the dimensionality of the space of action potential waveforms, and then performs clustering analysis to associate specific action potentials with individual neurons.

PCA as a dimension reduction technique is particularly suited to detect coordinated activities of large neuronal ensembles. It has been used in determining collective variables, that is, order parameters, during phase transitions in the brain. [55]

Relation with other methods

Correspondence analysis

Correspondence analysis (CA) was developed by Jean-Paul Benzécri [56] 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-squared statistic associated to this table into orthogonal factors. [57] Because CA is a descriptive technique, it can be applied to tables for which the chi-squared 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. [58]

Factor analysis

The above picture is an example of the difference between PCA and Factor Analysis. In the top diagram the "factor" (e.g., career path) represents the three observed variables (e.g., doctor, lawyer, teacher) whereas in the bottom diagram the observed variables (e.g., pre-school teacher, middle school teacher, high school teacher) are reduced into the component of interest (e.g., teacher). PCA versus Factor Analysis.jpg
The above picture is an example of the difference between PCA and Factor Analysis. In the top diagram the "factor" (e.g., career path) represents the three observed variables (e.g., doctor, lawyer, teacher) whereas in the bottom diagram the observed variables (e.g., pre-school teacher, middle school teacher, high school teacher) are reduced into the component of interest (e.g., teacher).

Principal component analysis creates variables that are linear combinations of the original variables. The new variables have the property that the variables are all orthogonal. The PCA transformation can be helpful as a pre-processing step before clustering. PCA is a variance-focused approach seeking to reproduce the total variable variance, in which components reflect both common and unique variance of the variable. PCA is generally preferred for purposes of data reduction (that is, translating variable space into optimal factor space) but not when the goal is to detect the latent construct or factors.

Factor analysis is similar to principal component analysis, in that factor analysis also involves linear combinations of variables. Different from PCA, factor analysis is a correlation-focused approach seeking to reproduce the inter-correlations among variables, in which the factors "represent the common variance of variables, excluding unique variance". [59] In terms of the correlation matrix, this corresponds with focusing on explaining the off-diagonal terms (that is, shared co-variance), while PCA focuses on explaining the terms that sit on the diagonal. However, as a side result, when trying to reproduce the on-diagonal terms, PCA also tends to fit relatively well the off-diagonal correlations. [10] :158 Results given by PCA and factor analysis are very similar in most situations, but this is not always the case, and there are some problems where the results are significantly different. Factor analysis is generally used when the research purpose is detecting data structure (that is, latent constructs or factors) or causal modeling. If the factor model is incorrectly formulated or the assumptions are not met, then factor analysis will give erroneous results. [60]

K-means clustering

It has been asserted that the relaxed solution of k-means clustering, specified by the cluster indicators, is given by the principal components, and the PCA subspace spanned by the principal directions is identical to the cluster centroid subspace. [61] [62] However, that PCA is a useful relaxation of k-means clustering was not a new result, [63] and it is straightforward to uncover counterexamples to the statement that the cluster centroid subspace is spanned by the principal directions. [64]

Non-negative matrix factorization

Fractional residual variance (FRV) plots for PCA and NMF; for PCA, the theoretical values are the contribution from the residual eigenvalues. In comparison, the FRV curves for PCA reaches a flat plateau where no signal are captured effectively; while the NMF FRV curves are declining continuously, indicating a better ability to capture signal. The FRV curves for NMF also converges to higher levels than PCA, indicating the less-overfitting property of NMF. Fractional Residual Variances comparison, PCA and NMF.pdf
Fractional residual variance (FRV) plots for PCA and NMF; for PCA, the theoretical values are the contribution from the residual eigenvalues. In comparison, the FRV curves for PCA reaches a flat plateau where no signal are captured effectively; while the NMF FRV curves are declining continuously, indicating a better ability to capture signal. The FRV curves for NMF also converges to higher levels than PCA, indicating the less-overfitting property of NMF.

Non-negative matrix factorization (NMF) is a dimension reduction method where only non-negative elements in the matrices are used, which is therefore a promising method in astronomy, [20] [21] [22] in the sense that astrophysical signals are non-negative. The PCA components are orthogonal to each other, while the NMF components are all non-negative and therefore constructs a non-orthogonal basis.

In PCA, the contribution of each component is ranked based on the magnitude of its corresponding eigenvalue, which is equivalent to the fractional residual variance (FRV) in analyzing empirical data. [18] For NMF, its components are ranked based only on the empirical FRV curves. [22] The residual fractional eigenvalue plots, that is, as a function of component number given a total of components, for PCA has a flat plateau, where no data is captured to remove the quasi-static noise, then the curves dropped quickly as an indication of over-fitting and captures random noise. [18] The FRV curves for NMF is decreasing continuously [22] when the NMF components are constructed sequentially, [21] indicating the continuous capturing of quasi-static noise; then converge to higher levels than PCA, [22] indicating the less over-fitting property of NMF.

Iconography of correlations

It is often difficult to interpret the principal components when the data include many variables of various origins, or when some variables are qualitative. This leads the PCA user to a delicate elimination of several variables. If observations or variables have an excessive impact on the direction of the axes, they should be removed and then projected as supplementary elements. In addition, it is necessary to avoid interpreting the proximities between the points close to the center of the factorial plane.

Iconography of correlations - Geochemistry of marine aerosols AirMerIconographyCorrelation.jpg
Iconography of correlations - Geochemistry of marine aerosols

The iconography of correlations, on the contrary, which is not a projection on a system of axes, does not have these drawbacks. We can therefore keep all the variables.

The principle of the diagram is to underline the "remarkable" correlations of the correlation matrix, by a solid line (positive correlation) or dotted line (negative correlation).

A strong correlation is not "remarkable" if it is not direct, but caused by the effect of a third variable. Conversely, weak correlations can be "remarkable". For example, if a variable Y depends on several independent variables, the correlations of Y with each of them are weak and yet "remarkable".


Sparse PCA

A particular disadvantage of PCA is that the principal components are usually linear combinations of all input variables. Sparse PCA overcomes this disadvantage by finding linear combinations that contain just a few input variables. It extends the classic method of principal component analysis (PCA) for the reduction of dimensionality of data by adding sparsity constraint on the input variables. Several approaches have been proposed, including

The methodological and theoretical developments of Sparse PCA as well as its applications in scientific studies were recently reviewed in a survey paper. [71]

Nonlinear PCA

Linear PCA versus nonlinear Principal Manifolds 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 cannot 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. Software is available for free non-commercial use. Elmap breastcancer wiki.png
Linear PCA versus nonlinear Principal Manifolds 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 cannot 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. Software is available for free non-commercial use.

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. Trevor Hastie expanded on this concept by proposing Principal curves [75] as the natural extension for the geometric interpretation of PCA, which explicitly constructs a manifold for data approximation followed by projecting the points onto it, as is illustrated by Fig. See also the elastic map algorithm and principal geodesic analysis. [76] Another popular generalization is kernel PCA, which corresponds to PCA performed in a reproducing kernel Hilbert space associated with a positive definite kernel.

In multilinear subspace learning, [77] PCA is generalized to multilinear PCA (MPCA) that extracts features directly from tensor representations. MPCA is solved by performing PCA in each mode of the tensor iteratively. MPCA has been applied to face recognition, gait recognition, etc. MPCA is further extended to uncorrelated MPCA, non-negative MPCA and robust MPCA.

N-way principal component analysis may be performed with models such as Tucker decomposition, PARAFAC, multiple factor analysis, co-inertia analysis, STATIS, and DISTATIS.

Robust PCA

While PCA finds the mathematically optimal method (as in minimizing the squared error), it is still sensitive to outliers in the data that produce large errors, something that the method tries to avoid in the first place. It is therefore common practice 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 [78] based on a weighted PCA increases robustness by assigning different weights to data objects based on their estimated relevancy.

Outlier-resistant variants of PCA have also been proposed, based on L1-norm formulations (L1-PCA). [5] [3]

Robust principal component analysis (RPCA) via decomposition in low-rank and sparse matrices is a modification of PCA that works well with respect to grossly corrupted observations. [79] [80] [81]

Similar techniques

Independent component analysis

Independent component analysis (ICA) is directed to similar problems as principal component analysis, but finds additively separable components rather than successive approximations.

Network component analysis

Given a matrix , it tries to decompose it into two matrices such that . A key difference from techniques such as PCA and ICA is that some of the entries of are constrained to be 0. Here is termed the regulatory layer. While in general such a decomposition can have multiple solutions, they prove that if the following conditions are satisfied :

  1. has full column rank
  2. Each column of must have at least zeroes where is the number of columns of (or alternatively the number of rows of ). The justification for this criterion is that if a node is removed from the regulatory layer along with all the output nodes connected to it, the result must still be characterized by a connectivity matrix with full column rank.
  3. must have full row rank.

then the decomposition is unique up to multiplication by a scalar. [82]

Discriminant analysis of principal components

Discriminant analysis of principal components (DAPC) is a multivariate method used to identify and describe clusters of genetically related individuals. Genetic variation is partitioned into two components: variation between groups and within groups, and it maximizes the former. Linear discriminants are linear combinations of alleles which best separate the clusters. Alleles that most contribute to this discrimination are therefore those that are the most markedly different across groups. The contributions of alleles to the groupings identified by DAPC can allow identifying regions of the genome driving the genetic divergence among groups [83] In DAPC, data is first transformed using a principal components analysis (PCA) and subsequently clusters are identified using discriminant analysis (DA).

A DAPC can be realized on R using the package Adegenet. (more info:

Software/source code

See also

Related Research Articles

Multivariate normal distribution Generalization of the one-dimensional normal distribution to higher dimensions

In probability theory and statistics, the multivariate normal distribution, multivariate Gaussian distribution, or joint normal distribution is a generalization of the one-dimensional (univariate) normal distribution to higher dimensions. One definition is that a random vector is said to be k-variate normally distributed if every linear combination of its k components has a univariate normal distribution. Its importance derives mainly from the multivariate central limit theorem. The multivariate normal distribution is often used to describe, at least approximately, any set of (possibly) correlated real-valued random variables each of which clusters around a mean value.

In probability theory and statistics, covariance is a measure of the joint variability of two random variables. If the greater values of one variable mainly correspond with the greater values of the other variable, and the same holds for the lesser values, the covariance is positive. In the opposite case, when the greater values of one variable mainly correspond to the lesser values of the other,, the covariance is negative. The sign of the covariance therefore shows the tendency in the linear relationship between the variables. The magnitude of the covariance is not easy to interpret because it is not normalized and hence depends on the magnitudes of the variables. The normalized version of the covariance, the correlation coefficient, however, shows by its magnitude the strength of the linear relation.

Covariance matrix Measure of covariance of components of a random vector

In probability theory and statistics, a covariance matrix is a square matrix giving the covariance between each pair of elements of a given random vector. Any covariance matrix is symmetric and positive semi-definite and its main diagonal contains variances.

Factor analysis is a statistical method used to describe variability among observed, correlated variables in terms of a potentially lower number of unobserved variables called factors. For example, it is possible that variations in six observed variables mainly reflect the variations in two unobserved (underlying) variables. Factor analysis searches for such joint variations in response to unobserved latent variables. The observed variables are modelled as linear combinations of the potential factors plus "error" terms, hence factor analysis can be thought of as a special case of errors-in-variables models.

In the mathematical discipline of linear algebra, a matrix decomposition or matrix factorization is a factorization of a matrix into a product of matrices. There are many different matrix decompositions; each finds use among a particular class of problems.

Nonlinear dimensionality reduction Summary of algorithms for nonlinear dimensionality reduction

High-dimensional data, meaning data that requires more than two or three dimensions to represent, can be difficult to interpret. One approach to simplification is to assume that the data of interest lies within lower-dimensional space. If the data of interest is of low enough dimension, the data can be visualised in the low-dimensional space.

Eigenface Set of eigenvectors used in the computer vision problem of human face recognition

An eigenface is the name given to a set of eigenvectors when used in the computer vision problem of human face recognition. The approach of using eigenfaces for recognition was developed by Sirovich and Kirby and used by Matthew Turk and Alex Pentland in face classification. The eigenvectors are derived from the covariance matrix of the probability distribution over the high-dimensional vector space of face images. The eigenfaces themselves form a basis set of all images used to construct the covariance matrix. This produces dimension reduction by allowing the smaller set of basis images to represent the original training images. Classification can be achieved by comparing how faces are represented by the basis set.

In statistics, sometimes the covariance matrix of a multivariate random variable is not known but has to be estimated. Estimation of covariance matrices then deals with the question of how to approximate the actual covariance matrix on the basis of a sample from the multivariate distribution. Simple cases, where observations are complete, can be dealt with by using the sample covariance matrix. The sample covariance matrix (SCM) is an unbiased and efficient estimator of the covariance matrix if the space of covariance matrices is viewed as an extrinsic convex cone in Rp×p; however, measured using the intrinsic geometry of positive-definite matrices, the SCM is a biased and inefficient estimator. In addition, if the random variable has a normal distribution, the sample covariance matrix has a Wishart distribution and a slightly differently scaled version of it is the maximum likelihood estimate. Cases involving missing data require deeper considerations. Another issue is the robustness to outliers, to which sample covariance matrices are highly sensitive.

In linear algebra, an eigenvector or characteristic vector of a linear transformation is a nonzero vector that changes at most by a scalar factor when that linear transformation is applied to it. The corresponding eigenvalue, often denoted by , is the factor by which the eigenvector is scaled.

In the field of multivariate statistics, kernel principal component analysis is an extension of principal component analysis (PCA) using techniques of kernel methods. Using a kernel, the originally linear operations of PCA are performed in a reproducing kernel Hilbert space.

A whitening transformation or sphering transformation is a linear transformation that transforms a vector of random variables with a known covariance matrix into a set of new variables whose covariance is the identity matrix, meaning that they are uncorrelated and each have variance 1. The transformation is called "whitening" because it changes the input vector into a white noise vector.

Two-dimensional singular-value decomposition (2DSVD) computes the low-rank approximation of a set of matrices such as 2D images or weather maps in a manner almost identical to SVD which computes the low-rank approximation of a single matrix.

The point distribution model is a model for representing the mean geometry of a shape and some statistical modes of geometric variation inferred from a training set of shapes.

Singular spectrum analysis Nonparametric spectral estimation method

In time series analysis, singular spectrum analysis (SSA) is a nonparametric spectral estimation method. It combines elements of classical time series analysis, multivariate statistics, multivariate geometry, dynamical systems and signal processing. Its roots lie in the classical Karhunen (1946)–Loève spectral decomposition of time series and random fields and in the Mañé (1981)–Takens (1981) embedding theorem. SSA can be an aid in the decomposition of time series into a sum of components, each having a meaningful interpretation. The name "singular spectrum analysis" relates to the spectrum of eigenvalues in a singular value decomposition of a covariance matrix, and not directly to a frequency domain decomposition.

In statistics, principal component regression (PCR) is a regression analysis technique that is based on principal component analysis (PCA). More specifically, PCR is used for estimating the unknown regression coefficients in a standard linear regression model.

Sparse principal component analysis is a specialised technique used in statistical analysis and, in particular, in the analysis of multivariate data sets. It extends the classic method of principal component analysis (PCA) for the reduction of dimensionality of data by introducing sparsity structures to the input variables.

Spike-triggered covariance (STC) analysis is a tool for characterizing a neuron's response properties using the covariance of stimuli that elicit spikes from a neuron. STC is related to the spike-triggered average (STA), and provides a complementary tool for estimating linear filters in a linear-nonlinear-Poisson (LNP) cascade model. Unlike STA, the STC can be used to identify a multi-dimensional feature space in which a neuron computes its response.

Common spatial pattern

Common spatial pattern (CSP) is a mathematical procedure used in signal processing for separating a multivariate signal into additive subcomponents which have maximum differences in variance between two windows.

Functional principal component analysis (FPCA) is a statistical method for investigating the dominant modes of variation of functional data. Using this method, a random function is represented in the eigenbasis, which is an orthonormal basis of the Hilbert space L2 that consists of the eigenfunctions of the autocovariance operator. FPCA represents functional data in the most parsimonious way, in the sense that when using a fixed number of basis functions, the eigenfunction basis explains more variation than any other basis expansion. FPCA can be applied for representing random functions, or in functional regression and classification.

In statistics, modes of variation are a continuously indexed set of vectors or functions that are centered at a mean and are used to depict the variation in a population or sample. Typically, variation patterns in the data can be decomposed in descending order of eigenvalues with the directions represented by the corresponding eigenvectors or eigenfunctions. Modes of variation provide a visualization of this decomposition and an efficient description of variation around the mean. Both in principal component analysis (PCA) and in functional principal component analysis (FPCA), modes of variation play an important role in visualizing and describing the variation in the data contributed by each eigencomponent. In real-world applications, the eigencomponents and associated modes of variation aid to interpret complex data, especially in exploratory data analysis (EDA).


  1. Barnett, T. P. & R. Preisendorfer. (1987). "Origins and levels of monthly and seasonal forecast skill for United States surface air temperatures determined by canonical correlation analysis". Monthly Weather Review. 115 (9): 1825. Bibcode:1987MWRv..115.1825B. doi: 10.1175/1520-0493(1987)115<1825:oaloma>;2 .
  2. Hsu, Daniel; Kakade, Sham M.; Zhang, Tong (2008). A spectral algorithm for learning hidden markov models. arXiv: 0811.4413 . Bibcode:2008arXiv0811.4413H.
  3. 1 2 Markopoulos, Panos P.; Kundu, Sandipan; Chamadia, Shubham; Pados, Dimitris A. (15 August 2017). "Efficient L1-Norm Principal-Component Analysis via Bit Flipping". IEEE Transactions on Signal Processing. 65 (16): 4252–4264. arXiv: 1610.01959 . Bibcode:2017ITSP...65.4252M. doi:10.1109/TSP.2017.2708023. S2CID   7931130.
  4. 1 2 Chachlakis, Dimitris G.; Prater-Bennette, Ashley; Markopoulos, Panos P. (22 November 2019). "L1-norm Tucker Tensor Decomposition". IEEE Access. 7: 178454–178465. arXiv: 1904.06455 . doi: 10.1109/ACCESS.2019.2955134 .
  5. 1 2 Markopoulos, Panos P.; Karystinos, George N.; Pados, Dimitris A. (October 2014). "Optimal Algorithms for L1-subspace Signal Processing". IEEE Transactions on Signal Processing. 62 (19): 5046–5058. arXiv: 1405.6785 . Bibcode:2014ITSP...62.5046M. doi:10.1109/TSP.2014.2338077. S2CID   1494171.
  6. Kanade, T.; Ke, Qifa (June 2005). Robust L1 Norm Factorization in the Presence of Outliers and Missing Data by Alternative Convex Programming. 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). Vol. 1. IEEE. p. 739. CiteSeerX . doi:10.1109/CVPR.2005.309. ISBN   978-0-7695-2372-9. S2CID   17144854.
  7. Pearson, K. (1901). "On Lines and Planes of Closest Fit to Systems of Points in Space". Philosophical Magazine. 2 (11): 559–572. doi:10.1080/14786440109462720.
  8. Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components. Journal of Educational Psychology , 24, 417–441, and 498–520.
    Hotelling, H (1936). "Relations between two sets of variates". Biometrika . 28 (3/4): 321–377. doi:10.2307/2333955. JSTOR   2333955.
  9. Stewart, G. W. (1993). "On the early history of the singular value decomposition". SIAM Review . 35 (4): 551–566. doi:10.1137/1035134.
  10. 1 2 3 4 5 Jolliffe, I. T. (2002). Principal Component Analysis. Springer Series in Statistics. New York: Springer-Verlag. doi:10.1007/b98835. ISBN   978-0-387-95442-4.
  11. Bengio, Y.; et al. (2013). "Representation Learning: A Review and New Perspectives". IEEE Transactions on Pattern Analysis and Machine Intelligence. 35 (8): 1798–1828. arXiv: 1206.5538 . doi:10.1109/TPAMI.2013.50. PMID   23787338. S2CID   393948.
  12. Forkman J., Josse, J., Piepho, H. P. (2019). "Hypothesis tests for principal component analysis when variables are standardized". Journal of Agricultural, Biological, and Environmental Statistics. 24 (2): 289–308. doi: 10.1007/s13253-019-00355-5 .{{cite journal}}: CS1 maint: multiple names: authors list (link)
  13. A. A. Miranda, Y. A. Le Borgne, and G. Bontempi. New Routes from Minimal Approximation Error to Principal Components, Volume 27, Number 3 / June, 2008, Neural Processing Letters, Springer
  14. Fukunaga, Keinosuke (1990). Introduction to Statistical Pattern Recognition. Elsevier. ISBN   978-0-12-269851-4.
  15. Alizadeh, Elaheh; Lyons, Samanthe M; Castle, Jordan M; Prasad, Ashok (2016). "Measuring systematic changes in invasive cancer cell shape using Zernike moments". Integrative Biology. 8 (11): 1183–1193. doi:10.1039/C6IB00100A. PMID   27735002.
  16. Leznik, M; Tofallis, C. 2005 Estimating Invariant Principal Components Using Diagonal Regression.
  17. Jonathon Shlens, A Tutorial on Principal Component Analysis.
  18. 1 2 3 Soummer, Rémi; Pueyo, Laurent; Larkin, James (2012). "Detection and Characterization of Exoplanets and Disks Using Projections on Karhunen-Loève Eigenimages". The Astrophysical Journal Letters. 755 (2): L28. arXiv: 1207.4197 . Bibcode:2012ApJ...755L..28S. doi:10.1088/2041-8205/755/2/L28. S2CID   51088743.
  19. Pueyo, Laurent (2016). "Detection and Characterization of Exoplanets using Projections on Karhunen Loeve Eigenimages: Forward Modeling". The Astrophysical Journal. 824 (2): 117. arXiv: 1604.06097 . Bibcode:2016ApJ...824..117P. doi:10.3847/0004-637X/824/2/117. S2CID   118349503.
  20. 1 2 Blanton, Michael R.; Roweis, Sam (2007). "K-corrections and filter transformations in the ultraviolet, optical, and near infrared". The Astronomical Journal. 133 (2): 734–754. arXiv: astro-ph/0606170 . Bibcode:2007AJ....133..734B. doi:10.1086/510127. S2CID   18561804.
  21. 1 2 3 Zhu, Guangtun B. (2016-12-19). "Nonnegative Matrix Factorization (NMF) with Heteroscedastic Uncertainties and Missing data". arXiv: 1612.06037 [astro-ph.IM].
  22. 1 2 3 4 5 6 Ren, Bin; Pueyo, Laurent; Zhu, Guangtun B.; Duchêne, Gaspard (2018). "Non-negative Matrix Factorization: Robust Extraction of Extended Structures". The Astrophysical Journal. 852 (2): 104. arXiv: 1712.10317 . Bibcode:2018ApJ...852..104R. doi:10.3847/1538-4357/aaa1f2. S2CID   3966513.
  23. "What are the Pros and cons of the PCA?". i2tutorials. September 1, 2019. Retrieved June 4, 2021.
  24. Abbott, Dean (May 2014). Applied Predictive Analytics. Wiley. ISBN   9781118727966.
  25. 1 2 Jiang, Hong; Eskridge, Kent M. (2000). "Bias in Principal Components Analysis Due to Correlated Observations". Conference on Applied Statistics in Agriculture. doi:10.4148/2475-7772.1247. ISSN   2475-7772.
  26. Linsker, Ralph (March 1988). "Self-organization in a perceptual network". IEEE Computer. 21 (3): 105–117. doi:10.1109/2.36. S2CID   1527671.
  27. Deco & Obradovic (1996). An Information-Theoretic Approach to Neural Computing. New York, NY: Springer. ISBN   9781461240167.
  28. Plumbley, Mark (1991). Information theory and unsupervised neural networks.Tech Note
  29. Geiger, Bernhard; Kubin, Gernot (January 2013). "Signal Enhancement as Minimization of Relevant Information Loss". Proc. ITG Conf. On Systems, Communication and Coding. arXiv: 1205.6935 . Bibcode:2012arXiv1205.6935G.
  30. "Engineering Statistics Handbook Section" . Retrieved 19 January 2015.
  31. A.A. Miranda, Y.-A. Le Borgne, and G. Bontempi. New Routes from Minimal Approximation Error to Principal Components, Volume 27, Number 3 / June, 2008, Neural Processing Letters, Springer
  32. Abdi. H. & Williams, L.J. (2010). "Principal component analysis". Wiley Interdisciplinary Reviews: Computational Statistics. 2 (4): 433–459. arXiv: 1108.4372 . doi:10.1002/wics.101. S2CID   122379222.
  33. "SAS/STAT(R) 9.3 User's Guide".
  34. eig function Matlab documentation
  35. MATLAB PCA-based Face recognition software
  36. Eigenvalues function Mathematica documentation
  37. 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.
  38. 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.
  39. Kramer, R. (1998). Chemometric Techniques for Quantitative Analysis. New York: CRC Press. ISBN   9780203909805.
  40. Andrecut, M. (2009). "Parallel GPU Implementation of Iterative PCA Algorithms". Journal of Computational Biology. 16 (11): 1593–1599. arXiv: 0811.1081 . doi:10.1089/cmb.2008.0221. PMID   19772385. S2CID   1362603.
  41. Warmuth, M. K.; Kuzmin, D. (2008). "Randomized online PCA algorithms with regret bounds that are logarithmic in the dimension" (PDF). Journal of Machine Learning Research. 9: 2287–2320.
  42. Kaplan, R.M., & Saccuzzo, D.P. (2010). Psychological Testing: Principles, Applications, and Issues. (8th ed.). Belmont, CA: Wadsworth, Cengage Learning.
  43. Shevky, Eshref; Williams, Marilyn (1949). The Social Areas of Los Angeles: Analysis and Typology. University of California Press.
  44. Flood, J (2000). Sydney divided: factorial ecology revisited. Paper to the APA Conference 2000, Melbourne,November and to the 24th ANZRSAI Conference, Hobart, December 2000.
  45. "Socio-Economic Indexes for Areas". Australian Bureau of Statistics. 2011. Retrieved 2022-05-05.
  46. Human Development Reports. "Human Development Index". United Nations Development Programme. Retrieved 2022-05-06.
  47. Novembre, John; Stephens, Matthew (2008). "Interpreting principal component analyses of spatial population genetic variation". Nat Genet. 40 (5): 646–49. doi:10.1038/ng.139. PMC   3989108 . PMID   18425127.
  48. DeSarbo, Wayne; Hausmann, Robert; Kukitz, Jeffrey (2007). "Restricted principal components analysis for marketing research". Journal of Marketing in Management. 2: 305–328 via Researchgate.
  49. Dutton, William H; Blank, Grant (2013). Cultures of the Internet: The Internet in Britain (PDF). Oxford Internet Institute. p. 6.
  50. Flood, Joe (2008). "Multinomial Analysis for Housing Careers Survey". Paper to the European Network for Housing Research Conference, Dublin. Retrieved 6 May 2022.
  51. The Pricing and Hedging of Interest Rate Derivatives: A Practical Guide to Swaps, J H M Darbyshire, 2016, ISBN   978-0995455511
  52. Giorgia Pasini (2017); Principal Component Analysis for Stock Portfolio Management. International Journal of Pure and Applied Mathematics. Volume 115 No. 1 2017, 153–167
  53. Libin Yang. An Application of Principal Component Analysis to Stock Portfolio Management. Department of Economics and Finance, University of Canterbury, January 2015.
  54. Brenner, N., Bialek, W., & de Ruyter van Steveninck, R.R. (2000).
  55. Jirsa, Victor; Friedrich, R; Haken, Herman; Kelso, Scott (1994). "A theoretical model of phase transitions in the human brain". Biological Cybernetics. 71 (1): 27–35. doi:10.1007/bf00198909. PMID   8054384. S2CID   5155075.
  56. Benzécri, J.-P. (1973). L'Analyse des Données. Volume II. L'Analyse des Correspondances. Paris, France: Dunod.
  57. Greenacre, Michael (1983). Theory and Applications of Correspondence Analysis. London: Academic Press. ISBN   978-0-12-299050-2.
  58. Le Roux; Brigitte and Henry Rouanet (2004). Geometric Data Analysis, From Correspondence Analysis to Structured Data Analysis. Dordrecht: Kluwer. ISBN   9781402022357.
  59. Timothy A. Brown. Confirmatory Factor Analysis for Applied Research Methodology in the social sciences. Guilford Press, 2006
  60. Meglen, R.R. (1991). "Examining Large Databases: A Chemometric Approach Using Principal Component Analysis". Journal of Chemometrics. 5 (3): 163–179. doi:10.1002/cem.1180050305. S2CID   120886184.
  61. H. Zha; C. Ding; M. Gu; X. He; H.D. Simon (Dec 2001). "Spectral Relaxation for K-means Clustering" (PDF). Neural Information Processing Systems Vol.14 (NIPS 2001): 1057–1064.
  62. Chris Ding; Xiaofeng He (July 2004). "K-means Clustering via Principal Component Analysis" (PDF). Proc. Of Int'l Conf. Machine Learning (ICML 2004): 225–232.
  63. Drineas, P.; A. Frieze; R. Kannan; S. Vempala; V. Vinay (2004). "Clustering large graphs via the singular value decomposition" (PDF). Machine Learning. 56 (1–3): 9–33. doi: 10.1023/b:mach.0000033113.59016.96 . S2CID   5892850 . Retrieved 2012-08-02.
  64. Cohen, M.; S. Elder; C. Musco; C. Musco; M. Persu (2014). Dimensionality reduction for k-means clustering and low rank approximation (Appendix B). arXiv: 1410.6801 . Bibcode:2014arXiv1410.6801C.
  65. Hui Zou; Trevor Hastie; Robert Tibshirani (2006). "Sparse principal component analysis" (PDF). Journal of Computational and Graphical Statistics . 15 (2): 262–286. CiteSeerX . doi:10.1198/106186006x113430. S2CID   5730904.
  66. Alexandre d'Aspremont; Laurent El Ghaoui; Michael I. Jordan; Gert R. G. Lanckriet (2007). "A Direct Formulation for Sparse PCA Using Semidefinite Programming" (PDF). SIAM Review . 49 (3): 434–448. arXiv: cs/0406021 . doi:10.1137/050645506. S2CID   5490061.
  67. Michel Journee; Yurii Nesterov; Peter Richtarik; Rodolphe Sepulchre (2010). "Generalized Power Method for Sparse Principal Component Analysis" (PDF). Journal of Machine Learning Research . 11: 517–553. arXiv: 0811.4724 . Bibcode:2008arXiv0811.4724J. CORE Discussion Paper 2008/70.
  68. Peter Richtarik; Martin Takac; S. Damla Ahipasaoglu (2012). "Alternating Maximization: Unifying Framework for 8 Sparse PCA Formulations and Efficient Parallel Codes". arXiv: 1212.4137 [stat.ML].
  69. Baback Moghaddam; Yair Weiss; Shai Avidan (2005). "Spectral Bounds for Sparse PCA: Exact and Greedy Algorithms" (PDF). Advances in Neural Information Processing Systems. Vol. 18. MIT Press.
  70. Yue Guan; Jennifer Dy (2009). "Sparse Probabilistic Principal Component Analysis" (PDF). Journal of Machine Learning Research Workshop and Conference Proceedings . 5: 185.
  71. Hui Zou; Lingzhou Xue (2018). "A Selective Overview of Sparse Principal Component Analysis". Proceedings of the IEEE . 106 (8): 1311–1320. doi: 10.1109/JPROC.2018.2846588 .
  72. A. N. Gorban, A. Y. Zinovyev, Principal Graphs and Manifolds, 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.
  73. 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. (2005). "Gene expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer". The Lancet . 365 (9460): 671–679. doi:10.1016/S0140-6736(05)17947-1. PMID   15721472. S2CID   16358549. Data online
  74. Zinovyev, A. "ViDaExpert – Multidimensional Data Visualization Tool". Institut Curie . Paris. (free for non-commercial use)
  75. Hastie, T.; Stuetzle, W. (June 1989). "Principal Curves" (PDF). Journal of the American Statistical Association . 84 (406): 502–506. doi:10.1080/01621459.1989.10478797.
  76. A.N. Gorban, B. Kegl, D.C. Wunsch, A. Zinovyev (Eds.), Principal Manifolds for Data Visualisation and Dimension Reduction, LNCSE 58, Springer, Berlin – Heidelberg – New York, 2007. ISBN   978-3-540-73749-0
  77. Lu, Haiping; Plataniotis, K.N.; Venetsanopoulos, A.N. (2011). "A Survey of Multilinear Subspace Learning for Tensor Data" (PDF). Pattern Recognition. 44 (7): 1540–1551. Bibcode:2011PatRe..44.1540L. doi:10.1016/j.patcog.2011.01.004.
  78. Kriegel, H. P.; Kröger, P.; Schubert, E.; Zimek, A. (2008). A General Framework for Increasing the Robustness of PCA-Based Correlation Clustering Algorithms. Scientific and Statistical Database Management. Lecture Notes in Computer Science. Vol. 5069. pp. 418–435. CiteSeerX . doi:10.1007/978-3-540-69497-7_27. ISBN   978-3-540-69476-2.
  79. Emmanuel J. Candes; Xiaodong Li; Yi Ma; John Wright (2011). "Robust Principal Component Analysis?". Journal of the ACM. 58 (3): 11. arXiv: 0912.3599 . doi:10.1145/1970392.1970395. S2CID   7128002.
  80. T. Bouwmans; E. Zahzah (2014). "Robust PCA via Principal Component Pursuit: A Review for a Comparative Evaluation in Video Surveillance". Computer Vision and Image Understanding. 122: 22–34. doi:10.1016/j.cviu.2013.11.009.
  81. T. Bouwmans; A. Sobral; S. Javed; S. Jung; E. Zahzah (2015). "Decomposition into Low-rank plus Additive Matrices for Background/Foreground Separation: A Review for a Comparative Evaluation with a Large-Scale Dataset". Computer Science Review. 23: 1–71. arXiv: 1511.01245 . Bibcode:2015arXiv151101245B. doi:10.1016/j.cosrev.2016.11.001. S2CID   10420698.
  82. Liao, J. C.; Boscolo, R.; Yang, Y.-L.; Tran, L. M.; Sabatti, C.; Roychowdhury, V. P. (2003). "Network component analysis: Reconstruction of regulatory signals in biological systems". Proceedings of the National Academy of Sciences. 100 (26): 15522–15527. Bibcode:2003PNAS..10015522L. doi: 10.1073/pnas.2136632100 . PMC   307600 . PMID   14673099.
  83. Liao, T.; Jombart, S.; Devillard, F.; Balloux (2010). "Discriminant analysis of principal components: a new method for the analysis of genetically structured populations". BMC Genetics. 11: 11:94. doi:10.1186/1471-2156-11-94. PMC   2973851 . PMID   20950446.
  84. "Principal Components Analysis". Institute for Digital Research and Education. UCLA. Retrieved 29 May 2018.

Further reading