Pascal Sarda

Functional linear model

In this paper, we study a regression model in which explanatory variables are sampling points of a continuous-time process. We propose an estimator of regression by means of a Functional Principal Component Analysis analogous to the one introduced by Bosq [(1991) NATO, ASI Series, pp. 509-529] in the case of Hilbertian AR processes. Both convergence in probability and almost sure convergence of this estimator are stated.

Nonparametric curve estimation from time series

Because of the sheer size and scope of the plastics industry, the title Developments in Plastics Technology now covers an incredibly wide range of subjects or topics. No single volume can survey the whole field in any depth and what follows is, therefore, a series of chapters on selected topics. The topics were selected by us, the editors, because of their immediate relevance to the plastics industry. When one considers the advancements of the plastics processing machinery (in terms of its speed of operation and conciseness of control), it was felt that several chapters should be included which related to the types of control systems used and the correct usage of hydraulics. The importance of using cellular, rubber-modified and engineering-type plastics has had a major impact on the plastics industry and therefore a chapter on each of these subjects has been included. The two remaining chapters are on the characterisation and behaviour of polymer structures, both subjects again being of current academic or industrial interest. Each of the contributions was written by a specialist in that field and to them all, we, the editors, extend our heartfelt thanks, as writing a contribution for a book such as this, while doing a full-time job, is no easy task.

Testing Hypotheses in The Functional Linear Model

The functional linear model with scalar response is a regression model where the predictor is a random function defined on some compact set of ℝ and the response is scalar. The response is modelled as Y=Ψ(X)+ɛ, where Ψ is some linear continuous operator defined on the space of square integrable functions and valued in ℝ. The random input X is independent from the noise ɛ. In this paper, we are interested in testing the null hypothesis of no effect, that is, the nullity of Ψ restricted to the Hilbert space generated by the random variable X. We introduce two test statistics based on the norm of the empirical cross‐covariance operator of (X,Y). The first test statistic relies on a χ2 approximation and we show the asymptotic normality of the second one under appropriate conditions on the covariance operator of X. The test procedures can be applied to check a given relationship between X and Y. The method is illustrated through a simulation study.

Smoothing splines estimators in functional linear regression with errors-in-variables

The total least squares method is generalized in the context of the functional linear model. A smoothing splines estimator of the functional coefficient of the model is first proposed without noise in the covariates and an asymptotic result for this estimator is obtained. Then, this estimator is adapted to the case where the covariates are noisy and an upper bound for the convergence speed is also derived. The estimation procedure is evaluated by means of simulations.

Quantile regression when the covariates are functions

This article deals with a linear model of regression on quantiles when the explanatory variable takes values in some functional space and the response is scalar. We propose a spline estimator of the functional coefficient that minimizes a penalized L 1 type criterion. Then, we study the asymptotic behavior of this estimator. The penalization is of primary importance to get existence and convergence.

Local data-driven bandwidth choice for density estimation

Abstract In the setting of kernel density estimation a local adaptation of Rudemo (1982) cross-validation technique leads to data-driven locally selected bandwidths. This method provides the estimate for which Mean Integrated Squared Error on some interval is smaller than that of the classical kernel estimate based on the global optimal bandwidth. As a by-product, extensions for local quadratic errors of asymptotic results of Marron and Hardle (1986) and Marron (1987) are stated.

Functional Linear Regression with Functional Response: Application to Prediction of Electricity Consumption

Functional linear regression model linking observations of a functional response variable with measurements of an explanatory functional variable is considered. The slope function is estimated with a tensor product splines. Some computational issues are addressed by means of a simulation study. This model serves to analyze a real data set concerning electricity consumption in Sardinia. The interest lies in predicting either incoming weekend or incoming weekdays consumption curves if actual weekdays consumption is known.

Testing for No Effect in Functional Linear Regression Models, Some Computational Approaches

Abstract The functional linear regression model is a regression model where the link between the response (a scalar) and the predictor (a random function) is expressed as an inner product between a functional coefficient and the predictor. Our aim is to test at first for no effect of the model, i.e., the nullity of the functional coefficient. A fully automatic permutation test based on the cross covariance operator of the predictor and the response is proposed. The model can be, in an obvious way, extended to the case of several functional predictors. When testing for no effect of some covariate on the response the permutation test is no longer valid. An alternative pseudo-likelihood ratio test statistic is then introduced. The procedure can be applied in some way to test partial nullity of a functional coefficient. All procedures are illustrated and compared by means of simulation studies.

Factor models and variable selection in high-dimensional regression analysis

The paper considers linear regression problems where the number of predictor variables is possibly larger than the sample size. The basic motivation of the study is to combine the points of view of model selection and functional regression by using a factor approach: it is assumed that the predictor vector can be decomposed into a sum of two uncorrelated random components reflecting common factors and specific variabilities of the explanatory variables. It is shown that the traditional assumption of a sparse vector of parameters is restrictive in this context. Common factors may possess a significant influence on the response variable which cannot be captured by the specific effects of a small number of individual variables. We therefore propose to include principal components as additional explanatory variables in an augmented regression model. We give finite sample inequalities for estimates of these components. It is then shown that model selection procedures can be used to estimate the parameters of the augmented model, and we derive theoretical properties of the estimators. Finite sample performance is illustrated by a simulation study.

Electricity consumption prediction with functional linear regression using spline estimators

A functional linear regression model linking observations of a functional response variable with measurements of an explanatory functional variable is considered. This model serves to analyse a real data set describing electricity consumption in Sardinia. The interest lies in predicting either oncoming weekends’ or oncoming weekdays’ consumption, provided actual weekdays’ consumption is known. A B-spline estimator of the functional parameter is used. Selected computational issues are addressed as well.