Functional data analysis

Functional data analysis (FDA) is a branch of statistics that analyses data providing information about curves, surfaces or anything else varying over a
continuum. In its most general form, under an FDA framework, each sample element of functional data is considered to be a random function. The physical
continuum over which these functions are defined is often time, but may also be spatial location, wavelength, probability, etc. Intrinsically, functional data are
infinite dimensional. The high intrinsic dimensionality of these data brings challenges for theory as well as computation, where these challenges vary with how the
functional data were sampled. However, the high or infinite dimensional structure of the data is a rich source of information and there are many interesting
challenges for research and data analysis.

History
Functional data analysis has roots going back to work by Grenander and Karhunen in the 1940s and 1950s.[1][2][3][4] They considered the decomposition of
square-integrable continuous time stochastic process into eigencomponents, now known as the Karhunen-Loève decomposition. A rigorous analysis of functional
principal components analysis was done in the 1970s by Kleffe, Dauxois and Pousse including results about the asymptotic distribution of the eigenvalues.[5][6]
More recently in the 1990s and 2000s the field has focused more on applications and understanding the effects of dense and sparse observations schemes. The
term "Functional Data Analysis" was coined by James O. Ramsay.[7]

Mathematical formalism
Random functions can be viewed as random elements taking values in a Hilbert space, or as a stochastic process. The former is mathematically convenient,
whereas the latter is somewhat more suitable from an applied perspective. These two approaches coincide if the random functions are continuous and a condition
called mean-squared continuity is satisfied.[8]

Hilbertian random variables

In the Hilbert space viewpoint, one considers an -valued random element , where is a separable Hilbert space such as the space of square-integrable

functions . Under the integrability condition that , one can define the mean of as the unique element
satisfying

This formulation is the Pettis integral but the mean can also be defined as Bochner integral . Under the integrability condition that is finite, the
covariance operator of is a linear operator that is uniquely defined by the relation

or, in tensor form, . The spectral theorem allows to decompose as the Karhunen-Loève decomposition

where are eigenvectors of , corresponding to the nonnegative eigenvalues of , in a non-increasing order. Truncating this infinite series to a finite order
underpins functional principal component analysis.

Stochastic processes

The Hilbertian point of view is mathematically convenient, but abstract; the above considerations do not necessarily even view as a function at all, since
common choices of like and Sobolev spaces consist of equivalence classes, not functions. The stochastic process perspective views as a collection
of random variables

indexed by the unit interval (or more generally interval ). The mean and covariance functions are defined in a pointwise manner as

(if for all ).

Under the mean square continuity, and are continuous functions and then the covariance function defines a covariance operator given b

   
(1)

The spectral theorem applies to , yielding eigenpairs , so that in tensor product notation writes
Moreover, since is continuous for all , all the are continuous. Mercer's theorem then states that

Finally, under the extra assumption that has continuous sample paths, namely that with probability one, the random function is continuous, the
Karhunen-Loève expansion above holds for and the Hilbert space machinery can be subsequently applied. Continuity of sample paths can be shown using
Kolmogorov continuity theorem.

Functional data designs

Functional data are considered as realizations of a stochastic process that is an process on a bounded and closed interval with mean
function and covariance function . The realizations of the process for the i-th subject is , and the sample is
assumed to consist of independent subjects. The sampling schedule may vary across subjects, denoted as for the i-th subject. The corresponding i-
th observation is denoted as , where . In addition, the measurement of is assumed to have random noise with
and , which are independent across and .

1. Fully observed functions without noise at arbitrarily dense grid

Measurements available for all

Often unrealistic but mathematically convenient.

Real life example: Tecator spectral data.[7]

2. Densely sampled functions with noisy measurements (dense design)

Measurements , where are recorded on a regular grid,

, and applies to typical functional data.

Real life example: Berkeley Growth Study Data (https://rdrr.io/cran/fda/man/growth.html#heading-0) and Stock data (https://www.marketwatch.com/investing/stoc
k/live)

3. Sparsely sampled functions with noisy measurements (longitudinal data)

Measurements , where are random times and their number per subject is random and finite.

Real life example: CD4 count data for AIDS patients.[9]

Functional principal component analysis

Functional principal component analysis (FPCA) is the most prevalent tool in FDA, partly because FPCA facilitates dimension reduction of the inherently infinite-
dimensional functional data to finite-dimensional random vector of scores. More specifically, dimension reduction is achieved by expanding the underlying
observed random trajectories in a functional basis consisting of the eigenfunctions of the covariance operator on . Consider the covariance operator
as in (1), which is a compact operator on Hilbert space.

By Mercer's theorem, the kernel of , i.e., the covariance function , has spectral decomposition , where the series

convergence is absolute and uniform, and are real-valued nonnegative eigenvalues in descending order with the corresponding orthonormal eigenfunctions

. By the Karhunen–Loève theorem, the FPCA expansion of an underlying random trajectory is , where

are the functional principal components (FPCs), sometimes referred to as scores. The Karhunen–Loève expansion facilitates

dimension reduction in the sense that the partial sum converges uniformly, i.e., as and thus the partial

sum with a large enough yields a good approximation to the infinite sum. Thereby, the information in is reduced from infinite dimensional to a -
dimensional vector with the approximated process ：

(2)
 

   
 

Other popular bases include spline, Fourier series and wavelet bases. Important applications of FPCA include the modes of variation and functional principal
component regression.

Functional linear regression models

Functional linear models can be viewed as an extension of the traditional multivariate linear models that associates vector responses with vector covariates. The
traditional linear model with scalar response and vector covariate can be expressed as

(3)
 

   
 

where denotes the inner product in Euclidean space, and denote the regression coefficients, and is a zero mean finite variance random
error (noise). Functional linear models can be divided into two types based on the responses.

Functional regression models with scalar response

Replacing the vector covariate and the coefficient vector in model (3) by a centered functional covariate and coefficient function
for and replacing the inner product in Euclidean space by that in Hilbert space , one arrives at the functional linear model

 
 

 
 
(4)

The simple functional linear model (4) can be extended to multiple functional covariates, , also including additional vector covariates ,
where , by

 
 

 
 
(5)

where is regression coefficient for , the domain of is , is the centered functional covariate given by , and is
regression coefficient function for , for . Models (4) and (5) have been studied extensively.[10][11][12]

Functional regression models with functional response

Consider a functional response on and multiple functional covariates , , . Two major models have been considered in this
setup.[13][7] One of these two models, generally referred to as functional linear model (FLM), can be written as:

   
 

where is the functional intercept, for , is a centered functional covariate on , is the corresponding
functional slopes with same domain, respectively, and is usually a random process with mean zero and finite variance.[13]
In this case, at any given time
, the value of , i.e., , depends on the entire trajectories of . Model (6) has been studied extensively.[14][15][16][17][18]

Function-on-scalar regression

In particular, taking as a constant function yields a special case of model (6)

which is a functional linear model with functional responses and scalar covariates.

Concurrent regression models

This model is given by,

 
 

   
(7)

where are functional covariates on , are the coefficient functions defined on the same interval and is usually assumed to be
a random process with mean zero and finite variance.[13] This model assumes that the value of depends on the current value of only and not the
history or future value. Hence, it is a "concurrent regression model", which is also referred as "varying-coefficient" model. Further, various
estimation methods have been proposed.[19][20][21][22][23][24]

Functional nonlinear regression models

Direct nonlinear extensions of the classical functional linear regression models (FLMs) still involve a linear predictor, but combine it with a nonlinear link function,
analogous to the idea of generalized linear model from the conventional linear model. Developments towards fully nonparametric regression models for functional
data encounter problems such as curse of dimensionality. In order to bypass the "curse" and the metric selection problem, we are motivated to consider nonlinear
functional regression models, which are subject to some structural constraints but do not overly infringe flexibility. One desires models that retain polynomial rates
of convergence, while being more flexible than, say, functional linear models. Such models are particularly useful when diagnostics for the functional linear model
indicate lack of fit, which is often encountered in real life situations. In particular, functional polynomial models, functional single and multiple index models and
functional additive models are three special cases of functional nonlinear regression models.

Functional polynomial regression models

Functional polynomial regression models may be viewed as a natural extension of the Functional Linear Models (FLMs) with scalar responses, analogous to
extending linear regression model to polynomial regression model. For a scalar response and a functional covariate with domain and the
corresponding centered predictor processes , the simplest and the most prominent member in the family of functional polynomial regression models is the
quadratic functional regression[25] given as follows,

where is the centered functional covariate, is a scalar coefficient, and are coefficient functions with domains and
, respectively. In addition to the parameter function β that the above functional quadratic regression model shares with the FLM, it also features a
parameter surface γ. By analogy to FLMs with scalar responses, estimation of functional polynomial models can be obtained through expanding both the centered
covariate and the coefficient functions and in an orthonormal basis.[25][26]

Functional single and multiple index models

A functional multiple index model is given as below, with symbols having their usual meanings as formerly described,

Here g represents an (unknown) general smooth function defined on a p-dimensional domain. The case yields a functional single index model while
multiple index models correspond to the case . However, for , this model is problematic due to curse of dimensionality. With and relatively
small sample sizes, the estimator given by this model often has large variance.[27][28]

Functional additive models (FAMs)

For a given orthonormal basis on , we can expand on the domain .

A functional linear model with scalar responses (see (3)) can thus be written as follows,

One form of FAMs is obtained by replacing the linear function of in the above expression ( i.e., ) by a general smooth function , analogous to the
extension of multiple linear regression models to additive models and is expressed as,

where satisfies for .[13][7] This constraint on the general smooth functions ensures identifiability in the sense that the estimates of
these additive component functions do not interfere with that of the intercept term . Another form of FAM is the continuously additive model,[29] expressed
as,

for a bivariate smooth additive surface which is required to satisfy for all , in order to ensure identifiability.

Generalized functional linear model

An obvious and direct extension of FLMs with scalar responses (see (3)) is to add a link function leading to a generalized functional linear model (GFLM)[30] in
analogy to the generalized linear model (GLM). The three components of the GFLM are:

1. Linear predictor ; [systematic component]

2. Variance function , where is the conditional mean; [random component]

3. Link function connecting the conditional mean and the linear predictor through . [systematic component]

Clustering and classification of functional data

For vector-valued multivariate data, k-means partitioning methods and hierarchical clustering are two main approaches. These classical clustering concepts for
vector-valued multivariate data have been extended to functional data. For clustering of functional data, k-means clustering methods are more popular than
hierarchical clustering methods. For k-means clustering on functional data, mean functions are usually regarded as the cluster centers. Covariance structures have
also been taken into consideration.[31] Besides k-means type clustering, functional clustering[32] based on mixture models is also widely used in clustering vector-
valued multivariate data and has been extended to functional data clustering.[33][34][35][36][37] Furthermore, Bayesian hierarchical clustering also plays an
important role in the development of model-based functional clustering.[38][39][40][41]

Functional classification assigns a group membership to a new data object either based on functional regression or functional discriminant analysis. Functional data
classification methods based on functional regression models use class levels as responses and the observed functional data and other covariates as predictors. For
regression based functional classification models, functional generalized linear models or more specifically, functional binary regression, such as functional logistic
regression for binary responses, are commonly used classification approaches. More generally, the generalized functional linear regression model based on the
FPCA approach is used.[42] Functional Linear Discriminant Analysis (FLDA) has also been considered as a classification method for functional
data.[43][44][45][46][47] Functional data classification involving density ratios has also been proposed.[48] A study of the asymptotic behavior of the proposed
classifiers in the large sample limit shows that under certain conditions the misclassification rate converges to zero, a phenomenon that has been referred to as
"perfect classification".[49]

Time warping

Motivations

In addition to amplitude variation,[50] time variation may also be assumed to

present in functional data. Time variation occurs when the subject-specific
timing of certain events of interest varies among subjects. One classical
example is the Berkeley Growth Study Data (https://rdrr.io/cran/fda/man/growt
h.html#heading-0),[51] where the amplitude variation is the growth rate and the
time variation explains the difference in children's biological age at which the
pubertal and the pre-pubertal growth spurt occurred. In the presence of time
variation, the cross-sectional mean function may not be an efficient estimate as
peaks and troughs are located randomly and thus meaningful signals may be
distorted or hidden.

Time warping, also known as curve registration,[52] curve alignment or time

synchronization, aims to identify and separate amplitude variation and time
variation. If both time and amplitude variation are present, then the observed
functional data can be modeled as , where
is a latent amplitude function and is a latent time warping
function that corresponds to a cumulative distribution function. The time Structures in cross-sectional mean destroyed if time variation is ignored. On the
warping functions are assumed to be invertible and to satisfy contrary, structures in cross-sectional mean is well-captured after restoring time
. variation.

The simplest case of a family of warping functions to specify phase variation is

linear transformation, that is , which warps the time of an underlying template function by subjected-specific shift and scale. More general class of
warping functions includes diffeomorphisms of the domain to itself, that is, loosely speaking, a class of invertible functions that maps the compact domain to itself
such that both the function and its inverse are smooth. The set of linear transformation is contained in the set of diffeomorphisms.[53] One challenge in time
warping is identifiability of amplitude and phase variation. Specific assumptions are required to break this non-identifiability.

Methods

Earlier approaches include dynamic time warping (DTW) used for applications such as speech recognition.[54] Another traditional method for time warping is
landmark registration,[55][56] which aligns special features such as peak locations to an average location. Other relevant warping methods include pairwise
warping,[57] registration using distance[53] and elastic warping.[58]

Dynamic time warping

The template function is determined through an iteration process, starting from cross-sectional mean, performing registration and recalculating the cross-sectional
mean for the warped curves, expecting convergence after a few iterations. DTW minimizes a cost function through dynamic programming. Problems of non-
smooth differentiable warps or greedy computation in DTW can be resolved by adding a regularization term to the cost function.

Landmark registration

Landmark registration (or feature alignment) assumes well-expressed features are present in all sample curves and uses the location of such features as a gold-
standard. Special features such as peak or trough locations in functions or derivatives are aligned to their average locations on the template function.[53] Then the
warping function is introduced through a smooth transformation from the average location to the subject-specific locations. A problem of landmark registration is
that the features may be missing or hard to identify due to the noise in the data.

Extensions
So far we considered scalar valued stochastic process, , defined on one dimensional time domain.

Multidimensional domain of
The domain of can be in , for example the data could be a sample of random surfaces.[59][60]

Multivariate stochastic process

The range set of the stochastic process may be extended from to [61][62][63] and further to nonlinear manifolds,[64] Hilbert spaces[65] and eventually to metric
spaces.[59]

Python packages
There are Python packages to work with functional data, and its representation, perform exploratory analysis, or preprocessing, and among other tasks such as
inference, classification, regression or clustering of functional data.

scikit-fda (https://github.com/GAA-UAM/scikit-fda)

R packages
Some packages can handle functional data under both dense and longitudinal designs.

fda (https://cran.r-project.org/web/packages/fda/index.html)[7]
refund (https://cran.r-project.org/web/packages/refund/index.html)
fdapace (https://cran.r-project.org/web/packages/fdapace/index.html)[17]
FDboost (https://cran.r-project.org/web/packages/FDboost/index.html)
classiFunc (https://www.rdocumentation.org/packages/classiFunc/versions/0.1.1)
fda.usc (https://cran.r-project.org/web/packages/fda.usc/index.html)
dtw (https://cran.r-project.org/web/packages/dtw/index.html)
fdasrvf (https://cran.r-project.org/web/packages/fdasrvf/index.html)[58]

See also
Functional principal component analysis
Karhunen–Loève theorem
Modes of variation
Functional regression
Generalized functional linear model
Stochastic processes
Lp space
Variance function

Further reading
Ramsay, J. O. and Silverman, B.W. (2005) Functional data analysis, 2nd ed., New York: Springer, ISBN 0-387-40080-X
Horvath, L. and Kokoszka, P. (2012) Inference for Functional Data with Applications, New York: Springer, ISBN 978-1-4614-3654-6
Hsing, T. and Eubank, R. (2015) Theoretical Foundations of Functional Data Analysis, with an Introduction to Linear Operators, Wiley series in
probability and statistics, John Wiley & Sons, Ltd, ISBN 978-0-470-01691-6
Morris, J. (2015) Functional Regression, Annual Review of Statistics and Its Application, Vol. 2, 321 - 359, https://doi.org/10.1146/annurev-
statistics-010814-020413
Wang et al. (2016) Functional Data Analysis, Annual Review of Statistics and Its Application, Vol. 3, 257-295, https://doi.org/10.1146/annurev-
statistics-041715-033624

Category:Regression analysis

References
1. Grenander, U. (1950). "Stochastic processes and statistical 5. Kleffe, J. (1973). "Principal components of random variables with
inference". Arkiv för Matematik. 1 (3): 195–277. values in a seperable hilbert space". Mathematische
Bibcode:1950ArM.....1..195G (https://ui.adsabs.harvard.edu/abs/195 Operationsforschung und Statistik. 4 (5): 391–406.
0ArM.....1..195G). doi:10.1007/BF02590638 (https://doi.org/10.100 doi:10.1080/02331887308801137 (https://doi.org/10.1080%2F0233
7%2FBF02590638). S2CID 120451372 (https://api.semanticschola 1887308801137).
r.org/CorpusID:120451372). 6. Dauxois, J; Pousse, A; Romain, Y. (1982). "Asymptotic theory for the
2. Rice, JA; Silverman, BW. (1991). "Estimating the mean and principal component analysis of a vector random function: Some
covariance structure nonparametrically when the data are curves". applications to statistical inference". Journal of Multivariate
Journal of the Royal Statistical Society. 53 (1): 233–243. Analysis. 12 (1): 136–154. doi:10.1016/0047-259X(82)90088-4 (http
3. Müller, HG. (2016). "Peter Hall, functional data analysis and random s://doi.org/10.1016%2F0047-259X%2882%2990088-4).
objects". Annals of Statistics. 44 (5): 1867–1887. doi:10.1214/16- 7. Ramsay, J; Silverman, BW. (2005). Functional Data Analysis, 2nd
AOS1492 (https://doi.org/10.1214%2F16-AOS1492). ed. Springer.
4. Karhunen, K (1946). Zur Spektraltheorie stochastischer Prozesse. 8. Hsing, T; Eubank, R (2015). Theoretical Foundations of Functional
Annales Academiae scientiarum Fennicae. Data Analysis, with an Introduction to Linear Operators. Wiley
Series in Probability and Statistics.
 9. Shi, M; Weiss, RE; Taylor, JMG. (1996). "An analysis of paediatric 25. Yao, F; Müller, HG. (2010). "Functional quadratic regression".
CD4 counts for acquired immune deficiency syndrome using Biometrika. 97 (1):49–64.
flexible random curves". Journal of the Royal Statistical Society. 26. Horváth, L; Reeder, R. (2013). "A test of significance in functional
Series C (Applied Statistics). 45 (2): 151–163. quadratic regression". Bernoulli. 19 (5A): 2120–2151.
10. Hilgert, N; Mas, A; Verzelen, N. (2013). "Minimax adaptive tests for doi:10.3150/12-BEJ446 (https://doi.org/10.3150%2F12-BEJ446).
the functional linear model". Annals of Statistics. 41 (2): 838–869. S2CID 88512527 (https://api.semanticscholar.org/CorpusID:885125
doi:10.1214/13-AOS1093 (https://doi.org/10.1214%2F13-AOS109 27).
3). S2CID 13119710 (https://api.semanticscholar.org/CorpusID:131 27. Chen, D; Hall, P; Müller HG. (2011). "Single and multiple index
19710). functional regression models with nonparametric link". The Annals
11. Kong, D; Xue, K; Yao, F; Zhang, HH. (2016). "Partially functional of Statistics. 39 (3):1720–1747.
linear regression in high dimensions". Biometrika. 103 (1): 147– 28. Jiang, CR; Wang JL. (2011). "Functional single index models for
159. doi:10.1093/biomet/asv062 (https://doi.org/10.1093%2Fbiome longitudinal data". he Annals of Statistics. 39 (1):362–388.
t%2Fasv062).
29. Müller HG; Wu Y; Yao, F. (2013). "Continuously additive models for
12. Horváth, L; Kokoszka, P. (2012). Inference for functional data with nonlinear functional regression". Biometrika. 100 (3): 607–622.
applications. Springer Series in Statistics. Springer-Verlag. doi:10.1093/biomet/ast004 (https://doi.org/10.1093%2Fbiomet%2Fa
13. Wang, JL; Chiou, JM; Müller, HG. (2016). "Functional data analysis" st004).
(https://zenodo.org/record/895750). Annual Review of Statistics and 30. Müller HG; Stadmüller, U. (2005). "Generalized Functional Linear
Its Application. 3 (1): 257–295. Bibcode:2016AnRSA...3..257W (http Models". The Annals of Statistics. 33 (2): 774–805.
s://ui.adsabs.harvard.edu/abs/2016AnRSA...3..257W). doi:10.1214/009053604000001156 (https://doi.org/10.1214%2F009
doi:10.1146/annurev-statistics-041715-033624 (https://doi.org/10.11 053604000001156).
46%2Fannurev-statistics-041715-033624). 31. Chiou, JM; Li, PL. (2007). "Functional clustering and identifying
14. Ramsay, JO; Dalzell, CJ. (1991). "Some tools for functional data substructures of longitudinal data". Journal of the Royal Statistical
analysis". Journal of the Royal Statistical Society, Series B Society, Series B (Statistical Methodology). 69 (4): 679–699.
(Methodological). 53 (3): 539–561. doi:10.1111/j.2517- doi:10.1111/j.1467-9868.2007.00605.x (https://doi.org/10.1111%2F
6161.1991.tb01844.x (https://doi.org/10.1111%2Fj.2517-6161.1991. j.1467-9868.2007.00605.x). S2CID 120883171 (https://api.semantic
tb01844.x). S2CID 118960346 (https://api.semanticscholar.org/Corp scholar.org/CorpusID:120883171).
usID:118960346). 32. Banfield, JD; Raftery, AE. (1993). "Model-based Gaussian and non-
15. Malfait, N; Ramsay, JO. (2003). "The historical functional linear Gaussian clustering". Biometrics. 49 (3): 803–821.
model". The Canadian Journal of Statistics. 31 (2): 115–128. doi:10.2307/2532201 (https://doi.org/10.2307%2F2532201).
doi:10.2307/3316063 (https://doi.org/10.2307%2F3316063). JSTOR 2532201 (https://www.jstor.org/stable/2532201).
JSTOR 3316063 (https://www.jstor.org/stable/3316063).
33. James, GM; Sugar, CA. (2003). "Clustering for sparsely sampled
S2CID 55092204 (https://api.semanticscholar.org/CorpusID:550922 functional data". Journal of the American Statistical Association. 98
04). (462): 397–408. doi:10.1198/016214503000189 (https://doi.org/10.1
16. He, G; Müller, HG; Wang, JL. (2003). "Functional canonical analysis 198%2F016214503000189). S2CID 9487422 (https://api.semantics
for square integrable stochastic processes". Journal of Multivariate cholar.org/CorpusID:9487422).
Analysis. 85 (1): 54–77. doi:10.1016/S0047-259X(02)00056-8 (http
34. Jacques, J; Preda, C. (2013). "Funclust: A curves clustering method
s://doi.org/10.1016%2FS0047-259X%2802%2900056-8).
using functional random variables density approximation" (https://ha
17. Yao, F; Müller, HG; Wang, JL. (2005). "Functional data analysis for l.archives-ouvertes.fr/hal-00628247v2/file/Paper-Funclust-V2-versio
sparse longitudinal data". Journal of the American Statistical nHAL.pdf) (PDF). Neurocomputing. 112: 164–171.
Association. 100 (470): 577–590. doi:10.1016/j.neucom.2012.11.042 (https://doi.org/10.1016%2Fj.ne
doi:10.1198/016214504000001745 (https://doi.org/10.1198%2F016 ucom.2012.11.042). S2CID 33591208 (https://api.semanticscholar.o
214504000001745). S2CID 1243975 (https://api.semanticscholar.or rg/CorpusID:33591208).
g/CorpusID:1243975).
35. Jacques, J; Preda, C. (2014). "Model-based clustering for
18. He, G; Müller, HG; Wang, JL; Yang, WJ. (2010). "Functional linear multivariate functional data". Computational Statistics & Data
regression via canonical analysis". Journal of Multivariate Analysis. Analysis. 71 (C): 92–106. doi:10.1016/j.csda.2012.12.004 (https://d
16 (3): 705–729. oi.org/10.1016%2Fj.csda.2012.12.004).
19. Fan, J; Zhang, W. (1999). "Statistical estimation in varying 36. Coffey, N; Hinde, J; Holian, E. (2014). "Clustering longitudinal
coefficient models". The Annals of Statistics. 27 (5): 1491–1518. profiles using P-splines and mixed effects models applied to time-
doi:10.1214/aos/1017939139 (https://doi.org/10.1214%2Faos%2F1 course gene expression data". Computational Statistics & Data
017939139). S2CID 16758288 (https://api.semanticscholar.org/Cor Analysis. 71 (C): 14–29. doi:10.1016/j.csda.2013.04.001 (https://doi.
pusID:16758288). org/10.1016%2Fj.csda.2013.04.001).
20. Wu, CO; Yu, KF. (2002). "Nonparametric varying-coefficient models 37. Heinzl, F; Tutz, G. (2014). "Clustering in linear-mixed models with a
for the analysis of longitudinal data". International Statistical group fused lasso penalty". Biometrical Journal. 56 (1): 44–68.
Review. 70 (3): 373–393. doi:10.1111/j.1751-5823.2002.tb00176.x doi:10.1002/bimj.201200111 (https://doi.org/10.1002%2Fbimj.2012
(https://doi.org/10.1111%2Fj.1751-5823.2002.tb00176.x). 00111). PMID 24249100 (https://pubmed.ncbi.nlm.nih.gov/2424910
S2CID 122007787 (https://api.semanticscholar.org/CorpusID:12200 0). S2CID 10969266 (https://api.semanticscholar.org/CorpusID:109
7787). 69266).
21. Huang, JZ; Wu, CO; Zhou, L. (2002). "Varying-coefficient models 38. Angelini, C; Canditiis, DD; Pensky, M. (2012). "Clustering time-
and basis function approximations for the analysis of repeated course microarray data using functional Bayesian infinite mixture
measurements". Biometrika. 89 (1): 111–128. model". Journal of Applied Statistics. 39 (1): 129–149.
doi:10.1093/biomet/89.1.111 (https://doi.org/10.1093%2Fbiomet%2 doi:10.1080/02664763.2011.578620 (https://doi.org/10.1080%2F02
F89.1.111). 664763.2011.578620). S2CID 8902492 (https://api.semanticschola
22. Huang, JZ; Wu, CO; Zhou, L. (2004). "Polynomial spline estimation r.org/CorpusID:8902492).
and inference for varying coefficient models with longitudinal data". 39. Rodríguez, A; Dunson, DB; Gelfand, AE. (2009). "Bayesian
Statistica Sinica. 14 (3): 763–788. nonparametric functional data analysis through density estimation"
23. Şentürk, D; Müller, HG. (2010). "Functional varying coefficient (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2650433).
models for longitudinal data". Journal of the American Statistical Biometrika. 96 (1): 149–162. doi:10.1093/biomet/asn054 (https://doi.
Association. 105 (491): 1256–1264. org/10.1093%2Fbiomet%2Fasn054). PMC 2650433 (https://www.n
doi:10.1198/jasa.2010.tm09228 (https://doi.org/10.1198%2Fjasa.20 cbi.nlm.nih.gov/pmc/articles/PMC2650433). PMID 19262739 (http
10.tm09228). S2CID 14296231 (https://api.semanticscholar.org/Cor s://pubmed.ncbi.nlm.nih.gov/19262739).
pusID:14296231). 40. Petrone, S; Guindani, M; Gelfand, AE. (2009). "Hybrid Dirichlet
24. Eggermont, PPB; Eubank, RL; LaRiccia, VN. (2010). "Convergence mixture models for functional data". Journal of the Royal Statistical
rates for smoothing spline estimators in varying coefficient models". Society. 71 (4): 755–782. doi:10.1111/j.1467-9868.2009.00708.x (ht
Journal of Statistical Planning and Inference. 140 (2): 369–381. tps://doi.org/10.1111%2Fj.1467-9868.2009.00708.x).
doi:10.1016/j.jspi.2009.06.017 (https://doi.org/10.1016%2Fj.jspi.200 S2CID 18638091 (https://api.semanticscholar.org/CorpusID:186380
9.06.017). 91).
41. Heinzl, F; Tutz, G. (2013). "Clustering in linear mixed models with 53. Marron, JS; Ramsay, JO; Sangalli, LM; Srivastava, A (2015).
approximate Dirichlet process mixtures using EM algorithm" (https:// "Functional data analysis of amplitude and phase variation".
epub.ub.uni-muenchen.de/23545/1/oa_23545.pdf) (PDF). Statistical Science. 30 (4): 468–484. arXiv:1512.03216 (https://arxiv.
Statistical Modelling. 13 (1): 41–67. org/abs/1512.03216). doi:10.1214/15-STS524 (https://doi.org/10.12
doi:10.1177/1471082X12471372 (https://doi.org/10.1177%2F14710 14%2F15-STS524). S2CID 55849758 (https://api.semanticscholar.
82X12471372). S2CID 11448616 (https://api.semanticscholar.org/C org/CorpusID:55849758).
orpusID:11448616). 54. Sakoe, H; Chiba, S. (1978). "Dynamic programming algorithm
42. Leng, X; Müller, HG. (2006). "Classification using functional data optimization for spoken word recognition". IEEE Transactions on
analysis for temporal gene expression data". Bioinformatics. 22 (1): Acoustics, Speech, and Signal Processing. 26: 43–49.
68–76. doi:10.1093/bioinformatics/bti742 (https://doi.org/10.1093%2 doi:10.1109/TASSP.1978.1163055 (https://doi.org/10.1109%2FTAS
Fbioinformatics%2Fbti742). PMID 16257986 (https://pubmed.ncbi.n SP.1978.1163055). S2CID 17900407 (https://api.semanticscholar.or
lm.nih.gov/16257986). g/CorpusID:17900407).
43. James, GM; Hastie, TJ. (2001). "Functional linear discriminant 55. Kneip, A; Gasser, T (1992). "Statistical tools to analyze data
analysis for irregularly sampled curves". Journal of the Royal representing a sample of curves". Annals of Statistics. 20 (3): 1266–
Statistical Society. 63 (3): 533–550. doi:10.1111/1467-9868.00297 1305. doi:10.1214/aos/1176348769 (https://doi.org/10.1214%2Fao
(https://doi.org/10.1111%2F1467-9868.00297). S2CID 16050693 (h s%2F1176348769).
ttps://api.semanticscholar.org/CorpusID:16050693). 56. Gasser, T; Kneip, A (1995). "Searching for structure in curve
44. Hall, P; Poskitt, DS; Presnell, B. (2001). "A Functional Data— sample". Journal of the American Statistical Association. 90 (432):
Analytic Approach to Signal Discrimination". Technometrics. 43 (1): 1179–1188.
1–9. doi:10.1198/00401700152404273 (https://doi.org/10.1198%2F 57. Tang, R; Müller, HG. (2008). "Pairwise curve synchronization for
00401700152404273). S2CID 21662019 (https://api.semanticschol functional data". Biometrika. 95 (4): 875–889.
ar.org/CorpusID:21662019). doi:10.1093/biomet/asn047 (https://doi.org/10.1093%2Fbiomet%2F
45. Ferraty, F; Vieu, P. (2003). "Curves discrimination: a nonparametric asn047).
functional approach". Computational Statistics & Data Analysis. 44 58. Anirudh, R; Turaga, P; Su, J; Srivastava, A (2015). "Elastic
(1–2): 161–173. doi:10.1016/S0167-9473(03)00032-X (https://doi.or functional coding of human actions: From vector-fields to latent
g/10.1016%2FS0167-9473%2803%2900032-X). variables". Proceedings of the IEEE Conference on Computer
46. Chang, C; Chen, Y; Ogden, RT. (2014). "Functional data Vision and Pattern Recognition: 3147–3155.
classification: a wavelet approach". Computational Statistics. 29 (6): 59. Dubey, P; Müller, HG (2021). "Modeling Time-Varying Random
1497–1513. doi:10.1007/s00180-014-0503-4 (https://doi.org/10.100 Objects and Dynamic Networks". Journal of the American Statistical
7%2Fs00180-014-0503-4). S2CID 120454400 (https://api.semantic Association. 117 (540): 2252–2267. arXiv:2104.04628 (https://arxiv.
scholar.org/CorpusID:120454400). org/abs/2104.04628). doi:10.1080/01621459.2021.1917416 (https://
47. Zhu, H; Brown, PJ; Morris, JS. (2012). "Robust Classification of doi.org/10.1080%2F01621459.2021.1917416). S2CID 233210300
Functional and Quantitative Image Data Using Functional Mixed (https://api.semanticscholar.org/CorpusID:233210300).
Models" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3443537). 60. Pigoli, D; Hadjipantelis, PZ; Coleman, JS; Aston, JAD (2017). "The
Biometrics. 68 (4): 1260–1268. doi:10.1111/j.1541- statistical analysis of acoustic phonetic data: exploring differences
0420.2012.01765.x (https://doi.org/10.1111%2Fj.1541-0420.2012.0 between spoken Romance languages". Journal of the Royal
1765.x). PMC 3443537 (https://www.ncbi.nlm.nih.gov/pmc/articles/P Statistical Society. Series C (Applied Statistics). 67 (5): 1130–1145.
MC3443537). PMID 22670567 (https://pubmed.ncbi.nlm.nih.gov/22
61. Happ, C; Greven, S (2018). "Multivariate Functional Principal
670567).
Component Analysis for Data Observed on Different (Dimensional)
48. Dai, X; Müller, HG; Yao, F. (2017). "Optimal Bayes classifiers for Domains". Journal of the American Statistical Association. 113
functional data and density ratios". Biometrika. 104 (3): 545–560. (522): 649–659. doi:10.1080/01621459.2016.1273115 (https://doi.or
arXiv:1605.03707 (https://arxiv.org/abs/1605.03707). g/10.1080%2F01621459.2016.1273115). S2CID 88521295 (https://
49. Delaigle, A; Hall, P (2012). "Achieving near perfect classification for api.semanticscholar.org/CorpusID:88521295).
functional data". Journal of the Royal Statistical Society. Series B 62. Chiou, JM; Yang, YF; Chen, YT (2014). "Multivariate functional
(Statistical Methodology). 74 (2): 267–286. doi:10.1111/j.1467- principal component analysis: a normalization approach". Statistica
9868.2011.01003.x (https://doi.org/10.1111%2Fj.1467-9868.2011.0 Sinica. 24: 1571–1596.
1003.x). ISSN 1369-7412 (https://www.worldcat.org/issn/1369-741
63. Carroll, C; Müller, HG; Kneip, A (2021). "Cross-component
2). S2CID 124261587 (https://api.semanticscholar.org/CorpusID:12
registration for multivariate functional data, with application to
4261587).
growth curves". Biometrics. 77 (3): 839–851. arXiv:1811.01429 (http
50. Wang, JL; Chiou, JM; Müller, HG. (2016). "Functional Data s://arxiv.org/abs/1811.01429). doi:10.1111/biom.13340 (https://doi.o
Analysis" (https://zenodo.org/record/895750). Annual Review of rg/10.1111%2Fbiom.13340). S2CID 220687157 (https://api.semanti
Statistics and Its Application. 3 (1): 257–295. cscholar.org/CorpusID:220687157).
Bibcode:2016AnRSA...3..257W (https://ui.adsabs.harvard.edu/abs/
64. Dai, X; Müller, HG (2018). "Principal component analysis for
2016AnRSA...3..257W). doi:10.1146/annurev-statistics-041715-
functional data on Riemannian manifolds and spheres". The Annals
033624 (https://doi.org/10.1146%2Fannurev-statistics-041715-0336
of Statistics. 46 (6B): 3334–3361. doi:10.1214/17-AOS1660 (https://
24).
doi.org/10.1214%2F17-AOS1660). S2CID 13671221 (https://api.se
51. Gasser, T; Müller, HG; Kohler, W; Molinari, L; Prader, A. (1984). manticscholar.org/CorpusID:13671221).
"Nonparametric regression analysis of growth curves". The Annals
65. Chen, K; Delicado, P; Müller, HG (2017). "Modelling function-
of Statistics. 12 (1): 210–229.
valued stochastic processes, with applications to fertility dynamics".
52. Ramsay, JO; Li, X. (1998). "Curve registration". Journal of the Royal Journal of the Royal Statistical Society. Series B (Statistical
Statistical Society, Series B. 60 (2): 351–363. doi:10.1111/1467- Methodology). 79 (1): 177–196. doi:10.1111/rssb.12160 (https://doi.
9868.00129 (https://doi.org/10.1111%2F1467-9868.00129). org/10.1111%2Frssb.12160). hdl:2117/126653 (https://hdl.handle.n
S2CID 17175587 (https://api.semanticscholar.org/CorpusID:171755 et/2117%2F126653). S2CID 13719492 (https://api.semanticscholar.
87). org/CorpusID:13719492).

Retrieved from "https://en.wikipedia.org/w/index.php?title=Functional_data_analysis&oldid=1135158527"