. Ceci-permet-donc, la seconde partie de ce chapitre, on a montré que les deux propriétés théoriques importantes, démontrées dans le cas direct, se transposaientàtransposaient`transposaientà l'approche par projection. On a en effet vu dans la section 6.3, que le perceptron multi-couches fonctionnel, basé sur uné etape de projectionétaitprojectionétait un approximateur universel. De plus, on a montré dans les sections 6.4.1 et 6.4.2 que l'estimation de ses paramètresparamètresétait consistante dans le cas d'une connaissance parfaite des fonctions d'entrée, comme dans le cas d'une connaissance empirique. On voit donc que d'un point de vue théorique

C. Abraham, P. Cornillon, E. Matzner-lober, and N. Molinari, Unsupervised Curve Clustering using B-Splines, Scandinavian Journal of Statistics, vol.78, issue.3, p.ENSAM?INRA?UM II?Montpellier, 2002.
DOI : 10.1111/1467-9469.00350
URL : http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.28.2845

W. K. Donald and . Andrews, Consistency in nonlinear econometric models : A generic uniform law of large numbers, Econometrica, vol.55, issue.6, pp.1465-1471, 1987.

R. Andrew and . Barron, Universal approximation bounds for superpositions of a sigmoidal function, IEEE Trans. Information Theory, vol.39, issue.3, pp.930-945, 1993.

P. Besse, Approximation spline de l'analyse en composantes principales d'une variable al??atoire hilbertienne, Annales de la facult?? des sciences de Toulouse Math??matiques, vol.12, issue.3, pp.329-349, 1991.
DOI : 10.5802/afst.729
URL : http://archive.numdam.org/article/AFST_1991_5_12_3_329_0.pdf

P. Besse, H. Cardot, and F. Ferraty, Simultaneous non-parametric regressions of unbalanced longitudinal data, Computational Statistics & Data Analysis, vol.24, issue.3, pp.255-270, 1997.
DOI : 10.1016/S0167-9473(96)00067-9

P. Besse, H. Cardot, and D. Stephenson, Autoregressive Forecasting of Some Functional Climatic Variations, Scandinavian Journal of Statistics, vol.27, issue.4, pp.673-688, 2000.
DOI : 10.1111/1467-9469.00215

P. Besse and J. Ramsay, Principal components analysis of sampled functions, Psychometrika, vol.78, issue.A, pp.285-311, 1986.
DOI : 10.1007/BF02293986

C. Philippe, H. Besse, and . Cardot, Approximation spline de la prévision d'un processus fonctionnel autorégressif d'ordre 1, Canadian Journal of Statistics, vol.24, pp.467-487, 1996.

C. Bishop and C. Legleye, Estimating conditional probability densities for periodic variables, 1995.
DOI : 10.1162/neco.1996.8.5.1123

D. Bosq, Modelization, non-parametric estimation and prediction for continuous time processes, Nonparametric Functional Estimation and Related Topics, Asi, pp.509-529, 1991.
DOI : 10.1007/978-94-011-3222-0_38

A. Babette, J. A. Brumback, and . Rice, Smoothing spline models for the analysis of nested and crossed samples of curves, J. Amer. Statist. Assoc, vol.93, pp.961-994, 1998.

H. Brézis, Analyse fonctionnelle : Théorie et applications. Dunod, 1994.

H. Cardot, Convergence du lissage spline de la pr??vision des processus autor??gressifs fonctionnels, Comptes Rendus de l'Acad??mie des Sciences - Series I - Mathematics, vol.326, issue.6, pp.755-758, 1998.
DOI : 10.1016/S0764-4442(98)80044-X

H. Cardot, Nonparametric estimation of smoothed principal components analysis of sampled noisy functions, Journal of Nonparametric Statistics, vol.53, issue.4, pp.503-538, 2000.
DOI : 10.1214/aos/1033066196

H. Cardot, F. Ferraty, A. Mas, and P. Sarda, Testing Hypotheses in the Functional Linear Model, Scandinavian Journal of Statistics, vol.29, issue.1, 2001.
DOI : 10.1016/S0167-7152(99)00036-X

H. Cardot, F. Ferraty, and P. Sarda, Functional linear model, Statistics & Probability Letters, vol.45, issue.1, pp.11-22, 1999.
DOI : 10.1016/S0167-7152(99)00036-X

H. Cardot, F. Ferraty, and P. Sarda, ??tude asymptotique d'un estimateur spline hybride pour le mod??le lin??aire fonctionnel, Comptes Rendus de l'Acad??mie des Sciences - Series I - Mathematics, vol.330, issue.6
DOI : 10.1016/S0764-4442(00)00207-X

T. Chen, A unified approach for neural network-like approximation of non-linear functionals, Neural Networks, vol.11, issue.6, pp.981-983, 1998.
DOI : 10.1016/S0893-6080(98)00075-6

T. Chen and H. Chen, Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems, IEEE Transactions on Neural Networks, vol.6, issue.4, pp.911-917, 1995.
DOI : 10.1109/72.392253

G. Cybenko, Approximation by superpositions of a sigmoidal function, mathematics of control, Signals and Systems, pp.303-314, 1989.

J. Dauxois, L. Ferré, and A. Yao, Un modèle semiparamétrique pour variables hilbertiennes, C. R. Acad. Sci. Paris
DOI : 10.1016/s0764-4442(01)02163-2

J. Dauxois and A. Pousse, Les analyses factorielles en calcul des probabilités et en statistiques : essai d'´ etude synthétique, 1976.

J. Dauxois, A. Pousse, and Y. Romain, Asymptotic theory for the principal component analysis of a vector random function: Some applications to statistical inference, Journal of Multivariate Analysis, vol.12, issue.1, 1982.
DOI : 10.1016/0047-259X(82)90088-4

T. Denoeux and R. Lengellé, Initializing back propagation networks with prototypes, Neural Networks, vol.6, issue.3, pp.351-363, 1993.
DOI : 10.1016/0893-6080(93)90003-F

J. C. Deville, M??thodes statistiques et num??riques de l'analyse harmonique, Annales de l'ins????, issue.15, 1974.
DOI : 10.2307/20075177

J. Fan, . Sheng-kuei, and . Lin, Test of Significance When Data are Curves, Journal of the American Statistical Association, vol.57, issue.443, pp.1007-1021, 1998.
DOI : 10.1016/S0002-9394(14)72213-5

J. Fan and J. Zhang, Two-step estimation of functional linear models with applications to longitudinal data, Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol.62, issue.2, pp.303-322, 2000.
DOI : 10.1111/1467-9868.00233

L. Ferré and A. Yao, Functional sliced inverse regression analysis, Statistics, vol.3, issue.6, 2002.
DOI : 10.2307/2291210

F. Ferraty and P. Vieu, Statistique fonctionnelle : Modèles de régression pour variables aléatoires uni, multi et infiniment dimensionnées, 2001.

F. Ferraty, Estimation non-paramétrique et discrimination de courbes, Actes deshuitì emes journées de la SFC, 2001.

F. Ferraty, A. Goia, and P. Vieu, R??gression non-param??trique pour des variables al??atoires fonctionnelles m??langeantes, Comptes Rendus Mathematique, vol.334, issue.3, pp.217-220, 2002.
DOI : 10.1016/S1631-073X(02)02248-3

F. Ferraty and P. Vieu, Dimension fractale et estimation de la r??gression dans des espaces vectoriels semi-norm??s, Comptes Rendus de l'Acad??mie des Sciences - Series I - Mathematics, vol.330, issue.2, pp.139-142, 2000.
DOI : 10.1016/S0764-4442(00)00140-3

F. Ferraty and P. Vieu, Statistique fonctionnelle : modèles de régression pour variables aléatoires uni, multi et infiniment dimensionnées, Laboratoire de statistique et probabilités, 2001.

R. Fletcher, Practical Methods of Optimization, 1987.
DOI : 10.1002/9781118723203

I. E. Frank and J. H. Friedman, A Statistical View of Some Chemometrics Regression Tools, Technometrics, vol.5, issue.2, pp.109-148, 1993.
DOI : 10.1080/00401706.1993.10485033

C. Gegout, B. Girau, and F. Rossi, Generic Back-Propagation in Arbitrary FeedForward Neural Networks, Int. Conf. on Artificial Neural Nets and Genetic Algorithms, pp.168-171, 1995.
DOI : 10.1007/978-3-7091-7535-4_45

T. Hastie, A. Buja, and E. R. Tibshirani, Penalized Discriminant Analysis, The Annals of Statistics, vol.23, issue.1, pp.73-102, 1995.
DOI : 10.1214/aos/1176324456
URL : http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.35.8378

T. Hastie and C. Mallows, A discussion of " a statistical view of some chemometrics regression tools " by i.e. frank and j.h. friedman, Technometrics, vol.35, pp.140-143, 1993.

T. Hastie and R. Tibshirani, Varying-coefficient models, Journal of the Royal Statistical Society B, vol.55, pp.757-796, 1993.

D. R. Hoover, J. A. Rice, C. O. Wu, and L. Yang, Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data, Biometrika, vol.85, issue.4, pp.809-822, 1998.
DOI : 10.1093/biomet/85.4.809

K. Hornik, Approximation capabilities of multilayer feedforward networks, Neural Networks, vol.4, issue.2, pp.251-257, 1991.
DOI : 10.1016/0893-6080(91)90009-T

K. Hornik, Some new results on neural network approximation, Neural Networks, vol.6, issue.8, pp.1069-1072, 1993.
DOI : 10.1016/S0893-6080(09)80018-X

K. Hornik, M. Stinchcombe, H. Et, and . White, Multilayer feedforward networks are universal approximators, Neural Networks, vol.2, issue.5, pp.359-366, 1989.
DOI : 10.1016/0893-6080(89)90020-8

K. and S. Ishwar, From decision trees to neural networks, 1990.

M. Gareth and . James, Generalized linear models with functional predictor variables, Journal of the Royal Statistical Society Series B, vol.64, pp.411-432, 2002.

M. Gareth, T. J. James, and . Hastie, Functional linear discriminant analysis of irregularly sampled curves, Journal of the Royal Statistical Society Series B, vol.63, pp.533-550, 2001.

M. Gareth, T. J. James, C. A. Hastie, and . Sugar, Principal component models for sparse functional data, Biometrika, vol.87, issue.3, pp.587-602, 2000.

M. Gareth, C. A. James, and . Sugar, Clustering for sparsely sampled functional data, 2002.

L. Lebart, A. Morineau, and M. Piron, Statistique exploratoire multidimensionnelle. Dunod, 3` eme edition, 1995.

M. Leshno, V. Y. Lin, A. Pinkus, and . Et-schocken-shimon, Multilayer feedforward networks with a nonpolynomial activation function can approximate any function, Neural Networks, vol.6, issue.6, pp.861-867, 1993.
DOI : 10.1016/S0893-6080(05)80131-5

S. Leurgans, R. Moyeed, and E. B. Silverman, Canonical correlation analysis when the data are curves, Journal of the Royal Statistical Society B, vol.55, issue.3, pp.725-740, 1993.

K. Li, Sliced Inverse Regression for Dimension Reduction, Journal of the American Statistical Association, vol.13, issue.414, pp.316-342, 1991.
DOI : 10.1214/aos/1176345514
URL : http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.452.161

F. A. Wessels-lodewyk and E. Barnard, Avoiding false local minima by proper initialization of connections, IEEE Transactions on Neural Networks, vol.3, issue.6, 1992.
DOI : 10.1109/72.165592

B. D. Marx and P. H. Eilers, Generalized linear regression on sampled signals with penalized likelihood, Statistical Modelling. Proceedings of the 11th International workshop on Statistical Modelling, 1996.

C. Michel-briand and Y. Escouffier, Segmentation d'un ensemble de courbes, pp.5-24, 1994.

J. Morel and S. Solimini, Variational methods in image segmentation, Progress in Nonlinear Differential Equations and Their Applications. Birkhäuser, 1995.
DOI : 10.1007/978-1-4684-0567-5

J. Steven, G. E. Nowlan, and . Hinton, Simplifying neural networks by soft weight-sharing, Neural Computation, vol.4, issue.4, pp.473-493, 1992.

S. Pezzulli and B. Silverman, On smoothed principal components analysis, BIBLIOGRAPHIE Computational Statistics, vol.8, pp.1-16, 1993.

E. Polak, Computational Methods in Optimization : A Unified Approach, 1971.

H. William, S. A. Press, W. T. Teukolsky, B. P. Vetterling, and . Flannery, Numerical Recipes : The Art of Scientific Computing

B. Pumo, Estimation et prévision de processus fonctionnels auto-régressifs : Application aux processusàprocessus`processusà temps continu, Thèse detroisì eme cycle, 1992.

J. Ramsay and B. Silverman, Functional Data Analysis, 1997.

A. John, B. Rice, and . Silverman, Estimating the mean and covariance structure nonparametrically when the data are curves, Journal of the Royal Statistical Society B, vol.53, issue.1, pp.233-243, 1991.

A. John, C. O. Rice, and . Wu, Nonparametric mixed effects models for unequally sampled noisy curves, Biometrics, vol.57, issue.1, pp.253-259, 2001.

F. Rossi and C. Gegout, Geometrical initialisation, parametrization and control of multilayer perceptrons : Application to function approximation, 1994.

W. Rudin, Real and complex Analysis, 1974.

I. W. Sandberg, Notes on weighted norms and network approximation of functionals IEEE Transactions on Circuits and Systems?I : Fundamental Theory and Applications, pp.600-601, 1996.

W. Irwin, L. Sandberg, and . Xu, Network approximation of input-output maps and functionals, Circuits Systems Signal Processing, vol.15, issue.6, pp.711-725, 1996.

B. Silverman, Incorporating parametric effects into functional principal components analysis Smooth functional principal components analysis by choice of norm, Journal of the Royal Statistical Society B The Annals of Statistics, vol.57, issue.241, pp.673-6891, 1995.

B. Maxwell and . Stinchcombe, Neural network approximation of continuous functionals and continuous functions on compactifications, Neural Networks, vol.12, issue.3, pp.467-477, 1999.

H. White, Learning in Artificial Neural Networks: A Statistical Perspective, Neural Computation, vol.36, issue.4, pp.425-464, 1989.
DOI : 10.2307/1912526

Y. Yu and D. Lambert, Fitting trees to functional data, with an application to time of day patterns, J. of Computational and Graphical Statistics, vol.8, pp.749-762, 2000.