Multivariate statistics

Last updated

Multivariate statistics is a subdivision of statistics encompassing the simultaneous observation and analysis of more than one outcome variable, i.e., multivariate random variables . Multivariate statistics concerns understanding the different aims and background of each of the different forms of multivariate analysis, and how they relate to each other. The practical application of multivariate statistics to a particular problem may involve several types of univariate and multivariate analyses in order to understand the relationships between variables and their relevance to the problem being studied.

Contents

In addition, multivariate statistics is concerned with multivariate probability distributions, in terms of both

  • how these can be used to represent the distributions of observed data;
  • how they can be used as part of statistical inference, particularly where several different quantities are of interest to the same analysis.

Certain types of problems involving multivariate data, for example simple linear regression and multiple regression, are not usually considered to be special cases of multivariate statistics because the analysis is dealt with by considering the (univariate) conditional distribution of a single outcome variable given the other variables.

Multivariate analysis

Multivariate analysis (MVA) is based on the principles of multivariate statistics. Typically, MVA is used to address situations where multiple measurements are made on each experimental unit and the relations among these measurements and their structures are important. [1] A modern, overlapping categorization of MVA includes: [1]

Multivariate analysis can be complicated by the desire to include physics-based analysis to calculate the effects of variables for a hierarchical "system-of-systems". Often, studies that wish to use multivariate analysis are stalled by the dimensionality of the problem. These concerns are often eased through the use of surrogate models, highly accurate approximations of the physics-based code. Since surrogate models take the form of an equation, they can be evaluated very quickly. This becomes an enabler for large-scale MVA studies: while a Monte Carlo simulation across the design space is difficult with physics-based codes, it becomes trivial when evaluating surrogate models, which often take the form of response-surface equations.

Types of analysis

Many different models are used in MVA, each with its own type of analysis:

  1. Multivariate analysis of variance (MANOVA) extends the analysis of variance to cover cases where there is more than one dependent variable to be analyzed simultaneously; see also Multivariate analysis of covariance (MANCOVA).
  2. Multivariate regression attempts to determine a formula that can describe how elements in a vector of variables respond simultaneously to changes in others. For linear relations, regression analyses here are based on forms of the general linear model. Some suggest that multivariate regression is distinct from multivariable regression, however, that is debated and not consistently true across scientific fields. [2]
  3. Principal components analysis (PCA) creates a new set of orthogonal variables that contain the same information as the original set. It rotates the axes of variation to give a new set of orthogonal axes, ordered so that they summarize decreasing proportions of the variation.
  4. Factor analysis is similar to PCA but allows the user to extract a specified number of synthetic variables, fewer than the original set, leaving the remaining unexplained variation as error. The extracted variables are known as latent variables or factors; each one may be supposed to account for covariation in a group of observed variables.
  5. Canonical correlation analysis finds linear relationships among two sets of variables; it is the generalised (i.e. canonical) version of bivariate [3] correlation.
  6. Redundancy analysis (RDA) is similar to canonical correlation analysis but allows the user to derive a specified number of synthetic variables from one set of (independent) variables that explain as much variance as possible in another (independent) set. It is a multivariate analogue of regression. [4]
  7. Correspondence analysis (CA), or reciprocal averaging, finds (like PCA) a set of synthetic variables that summarise the original set. The underlying model assumes chi-squared dissimilarities among records (cases).
  8. Canonical (or "constrained") correspondence analysis (CCA) for summarising the joint variation in two sets of variables (like redundancy analysis); combination of correspondence analysis and multivariate regression analysis. The underlying model assumes chi-squared dissimilarities among records (cases).
  9. Multidimensional scaling comprises various algorithms to determine a set of synthetic variables that best represent the pairwise distances between records. The original method is principal coordinates analysis (PCoA; based on PCA).
  10. Discriminant analysis, or canonical variate analysis, attempts to establish whether a set of variables can be used to distinguish between two or more groups of cases.
  11. Linear discriminant analysis (LDA) computes a linear predictor from two sets of normally distributed data to allow for classification of new observations.
  12. Clustering systems assign objects into groups (called clusters) so that objects (cases) from the same cluster are more similar to each other than objects from different clusters.
  13. Recursive partitioning creates a decision tree that attempts to correctly classify members of the population based on a dichotomous dependent variable.
  14. Artificial neural networks extend regression and clustering methods to non-linear multivariate models.
  15. Statistical graphics such as tours, parallel coordinate plots, scatterplot matrices can be used to explore multivariate data.
  16. Simultaneous equations models involve more than one regression equation, with different dependent variables, estimated together.
  17. Vector autoregression involves simultaneous regressions of various time series variables on their own and each other's lagged values.
  18. Principal response curves analysis (PRC) is a method based on RDA that allows the user to focus on treatment effects over time by correcting for changes in control treatments over time. [5]
  19. Iconography of correlations consists in replacing a correlation matrix by a diagram where the “remarkable” correlations are represented by a solid line (positive correlation), or a dotted line (negative correlation).

Dealing with incomplete data

It is very common that in an experimentally acquired set of data the values of some components of a given data point are missing. Rather than discarding the whole data point, it is common to "fill in" values for the missing components, a process called "imputation". [6]

Important probability distributions

There is a set of probability distributions used in multivariate analyses that play a similar role to the corresponding set of distributions that are used in univariate analysis when the normal distribution is appropriate to a dataset. These multivariate distributions are:

The Inverse-Wishart distribution is important in Bayesian inference, for example in Bayesian multivariate linear regression. Additionally, Hotelling's T-squared distribution is a multivariate distribution, generalising Student's t-distribution, that is used in multivariate hypothesis testing.

History

Anderson's 1958 textbook, An Introduction to Multivariate Statistical Analysis, [7] educated a generation of theorists and applied statisticians; Anderson's book emphasizes hypothesis testing via likelihood ratio tests and the properties of power functions: admissibility, unbiasedness and monotonicity. [8] [9]

MVA was formerly discussed solely in the context of statistical theories, due to the size and complexity of underlying datasets and its high computational consumption. With the dramatic growth of computational power, MVA now plays an increasingly important role in data analysis and has wide application in Omics fields.

Applications

Software and tools

There are an enormous number of software packages and other tools for multivariate analysis, including:

See also

Related Research Articles

The following outline is provided as an overview of and topical guide to statistics:

<span class="mw-page-title-main">Principal component analysis</span> Method of data analysis

Principal component analysis (PCA) is a linear dimensionality reduction technique with applications in exploratory data analysis, visualization and data preprocessing.

<span class="mw-page-title-main">Time series</span> Sequence of data points over time

In mathematics, a time series is a series of data points indexed in time order. Most commonly, a time series is a sequence taken at successive equally spaced points in time. Thus it is a sequence of discrete-time data. Examples of time series are heights of ocean tides, counts of sunspots, and the daily closing value of the Dow Jones Industrial Average.

<span class="mw-page-title-main">Regression analysis</span> Set of statistical processes for estimating the relationships among variables

In statistical modeling, regression analysis is a set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables. The most common form of regression analysis is linear regression, in which one finds the line that most closely fits the data according to a specific mathematical criterion. For example, the method of ordinary least squares computes the unique line that minimizes the sum of squared differences between the true data and that line. For specific mathematical reasons, this allows the researcher to estimate the conditional expectation of the dependent variable when the independent variables take on a given set of values. Less common forms of regression use slightly different procedures to estimate alternative location parameters or estimate the conditional expectation across a broader collection of non-linear models.

The general linear model or general multivariate regression model is a compact way of simultaneously writing several multiple linear regression models. In that sense it is not a separate statistical linear model. The various multiple linear regression models may be compactly written as

<span class="mw-page-title-main">Mathematical statistics</span> Branch of statistics

Mathematical statistics is the application of probability theory, a branch of mathematics, to statistics, as opposed to techniques for collecting statistical data. Specific mathematical techniques which are used for this include mathematical analysis, linear algebra, stochastic analysis, differential equations, and measure theory.

Linear discriminant analysis (LDA), normal discriminant analysis (NDA), or discriminant function analysis is a generalization of Fisher's linear discriminant, a method used in statistics and other fields, to find a linear combination of features that characterizes or separates two or more classes of objects or events. The resulting combination may be used as a linear classifier, or, more commonly, for dimensionality reduction before later classification.

In statistics, the generalized canonical correlation analysis (gCCA), is a way of making sense of cross-correlation matrices between the sets of random variables when there are more than two sets. While a conventional CCA generalizes principal component analysis (PCA) to two sets of random variables, a gCCA generalizes PCA to more than two sets of random variables. The canonical variables represent those common factors that can be found by a large PCA of all of the transformed random variables after each set underwent its own PCA.

This glossary of statistics and probability is a list of definitions of terms and concepts used in the mathematical sciences of statistics and probability, their sub-disciplines, and related fields. For additional related terms, see Glossary of mathematics and Glossary of experimental design.

In statistics, canonical analysis (from Ancient Greek: κανων bar, measuring rod, ruler) belongs to the family of regression methods for data analysis. Regression analysis quantifies a relationship between a predictor variable and a criterion variable by the coefficient of correlation r, coefficient of determination r2, and the standard regression coefficient β. Multiple regression analysis expresses a relationship between a set of predictor variables and a single criterion variable by the multiple correlation R, multiple coefficient of determination R2, and a set of standard partial regression weights β1, β2, etc. Canonical variate analysis captures a relationship between a set of predictor variables and a set of criterion variables by the canonical correlations ρ1, ρ2, ..., and by the sets of canonical weights C and D.

In mathematics, a univariate object is an expression, equation, function or polynomial involving only one variable. Objects involving more than one variable are multivariate. In some cases the distinction between the univariate and multivariate cases is fundamental; for example, the fundamental theorem of algebra and Euclid's algorithm for polynomials are fundamental properties of univariate polynomials that cannot be generalized to multivariate polynomials.

<span class="mw-page-title-main">Data transformation (statistics)</span>

In statistics, data transformation is the application of a deterministic mathematical function to each point in a data set—that is, each data point zi is replaced with the transformed value yi = f(zi), where f is a function. Transforms are usually applied so that the data appear to more closely meet the assumptions of a statistical inference procedure that is to be applied, or to improve the interpretability or appearance of graphs.

<span class="mw-page-title-main">Hydrological model</span>

A hydrologic model is a simplification of a real-world system that aids in understanding, predicting, and managing water resources. Both the flow and quality of water are commonly studied using hydrologic models.

<span class="mw-page-title-main">Bivariate analysis</span> Concept in statistical analysis

Bivariate analysis is one of the simplest forms of quantitative (statistical) analysis. It involves the analysis of two variables, for the purpose of determining the empirical relationship between them.

<span class="mw-page-title-main">Continuous or discrete variable</span> Types of quantitative variables in mathematics

In mathematics and statistics, a quantitative variable may be continuous or discrete if they are typically obtained by measuring or counting, respectively. If it can take on two particular real values such that it can also take on all real values between them, the variable is continuous in that interval. If it can take on a value such that there is a non-infinitesimal gap on each side of it containing no values that the variable can take on, then it is discrete around that value. In some contexts a variable can be discrete in some ranges of the number line and continuous in others.

A vine is a graphical tool for labeling constraints in high-dimensional probability distributions. A regular vine is a special case for which all constraints are two-dimensional or conditional two-dimensional. Regular vines generalize trees, and are themselves specializations of Cantor tree.

In statistics, linear regression is a statistical model which estimates the linear relationship between a scalar response and one or more explanatory variables. The case of one explanatory variable is called simple linear regression; for more than one, the process is called multiple linear regression. This term is distinct from multivariate linear regression, where multiple correlated dependent variables are predicted, rather than a single scalar variable. If the explanatory variables are measured with error then errors-in-variables models are required, also known as measurement error models.

Univariate is a term commonly used in statistics to describe a type of data which consists of observations on only a single characteristic or attribute. A simple example of univariate data would be the salaries of workers in industry. Like all the other data, univariate data can be visualized using graphs, images or other analysis tools after the data is measured, collected, reported, and analyzed.

The following outline is provided as an overview of and topical guide to machine learning:

References

  1. 1 2 Olkin, I.; Sampson, A. R. (2001-01-01), "Multivariate Analysis: Overview", in Smelser, Neil J.; Baltes, Paul B. (eds.), International Encyclopedia of the Social & Behavioral Sciences, Pergamon, pp. 10240–10247, ISBN   9780080430768 , retrieved 2019-09-02
  2. Hidalgo, B; Goodman, M (2013). "Multivariate or multivariable regression?". Am J Public Health. 103 (1): 39–40. doi:10.2105/AJPH.2012.300897. PMC   3518362 . PMID   23153131.
  3. Unsophisticated analysts of bivariate Gaussian problems may find useful a crude but accurate method of accurately gauging probability by simply taking the sum S of the N residuals' squares, subtracting the sum Sm at minimum, dividing this difference by Sm, multiplying the result by (N - 2) and taking the inverse anti-ln of half that product.
  4. Van Den Wollenberg, Arnold L. (1977). "Redundancy analysis an alternative for canonical correlation analysis". Psychometrika. 42 (2): 207–219. doi:10.1007/BF02294050.
  5. ter Braak, Cajo J.F. & Šmilauer, Petr (2012). Canoco reference manual and user's guide: software for ordination (version 5.0), p292. Microcomputer Power, Ithaca, NY.
  6. J.L. Schafer (1997). Analysis of Incomplete Multivariate Data. Chapman & Hall/CRC. ISBN   978-1-4398-2186-2.
  7. T.W. Anderson (1958) An Introduction to Multivariate Analysis, New York: Wiley ISBN   0471026409; 2e (1984) ISBN   0471889873; 3e (2003) ISBN   0471360910
  8. Sen, Pranab Kumar; Anderson, T. W.; Arnold, S. F.; Eaton, M. L.; Giri, N. C.; Gnanadesikan, R.; Kendall, M. G.; Kshirsagar, A. M.; et al. (June 1986). "Review: Contemporary Textbooks on Multivariate Statistical Analysis: A Panoramic Appraisal and Critique". Journal of the American Statistical Association . 81 (394): 560–564. doi:10.2307/2289251. ISSN   0162-1459. JSTOR   2289251.(Pages 560–561)
  9. Schervish, Mark J. (November 1987). "A Review of Multivariate Analysis". Statistical Science. 2 (4): 396–413. doi: 10.1214/ss/1177013111 . ISSN   0883-4237. JSTOR   2245530.
  10. CRAN has details on the packages available for multivariate data analysis

Further reading