Abstract
This article considers the problem of nonparametric estimation of the regression function \(r\) in a functional regression model \(Y = r(X) +\varepsilon \) with a scalar response Y, a functional explanatory variable X, and a second order stationary error process \(\varepsilon \). Under some specific criteria, we construct a local linear kernel estimator of \(r\) from functional random design with correlated errors. The exact rates of convergence of mean squared error of the constructed estimator are established for both short and long range dependent error processes. Simulation studies are conducted on the performance of the proposed simple local linear estimator. Examples of time series data are considered.
Similar content being viewed by others
References
Aneiros-Pérez G, Vieu P (2008) Nonparametric time series prediction: a semi-functional partial linear modeling. J Multivar Anal 99:834–857
Aneiros-Pérez G, Cao R, Vilar-Fernández J (2011) Functional methods for time series prediction: a nonparametric approach. J Forecast 30:377–392
Aneiros-Pérez G, Honová I, Husková M, Vieu P (2022) Special Issue on Functional data analysis and related fields. J Multivar Anal 189:104861
Baillo A, Grané A (2009) Locally modeled regression and functional data. J Multivar Anal 100:102–111
Barrientos-Marin J, Ferraty F, Vieu P (2010) Locally modeled regression and function data. J Nonparam Stat 22:617–632
Benhenni K, Ferraty F, Rachdi M, Vieu P (2007) Local smoothing regression with functional data. Comput Stat 22:353–369
Benhenni K, Hedli-Griche S, Rachdi M (2010) Estimation of the regression operator from functional fixed-design with correlated errors. J Multivar Anal 101:476–490
Benhenni K, Hedli-Griche S, Rachdi M (2017) Regression models with correlated errors based on functional random design. TEST 26:1–21
Benhenni K, Hajj-Hassan A, Su Y (2019) Local polynomial estimation of regression operators from functional data with correlated errors. J Multivar Anal 170:80–94
Beran J (1992) Statistical method for data with long range dependence. J Stat Sci 7:404–420
Beran J (1994) Statistics for long-memory processes. Chapman & Hall, New York
Berlinet A, Elamine A, Mas A (2011) Local linear regression for functional data. Ann Inst Stat Math 63:1047–1075
Boj E, Delicado D, Fortiana J (2004) Distance-based local linear regression for functional predictor. Comput Stat Data Anal 54:429–437
Cox DR (1984) Long-range dependent. In: David HA, David HT (eds) Statistics: an appraisal. Iowa State University Press, Iowa, pp 55–73
Demongeot J, Laksaci A, Madani F, Rachdi M (2010) Local linear estimation of the conditional density for functional data. C R Math Acad Sci Paris 348:931–934
Fan J, Gijbels I (1996) Local polynomial modeling and its applications. Chapman & Hall, London
Ferraty F (ed) (2011) Recent advances in functional data analysis and related topics. Physica-Verlag, Berlin
Ferraty F, Nagy S (2021) Scalar-on-function local linear regression and beyond. Biometrika 00:1–17
Ferraty F, Vieu P (2006) Nonparametric analysis for functional data: theory and practice. Springer, New York
Ferraty F, Rabhi A, Vieu P (2005) Conditional quantiles for dependent functional data with application to the climatic El Niño phenomenon. Sankhya 67:378–398
Ferraty F, Mas A, Vieu P (2007) Nonparametric regression on functional data: inference and practical aspects. Aust N Z J Stat 49:267–286
Horváth L, Kokoszka P (2012) Inference for functional data with applications. Springer, New York
Li WV, Shao QM (2001) Gaussian processes: inequalities, small ball probabilities and applications. In: Stochastic processes: theory and methods, Handbook of statistics, vol 19. North-Holland, Amsterdam
Masry E (2003) Local polynomial regression estimation with correlated errors. J Multivar Anal 86:330–359
Petersen A, Müler H-G (2019) Fréchet regression for random objects with Euckidean predictors. Ann Stat 49(2):691–719
Rachdi M, Vieu P (2007) Nonparametric regression for functional data: automatic smoothing parameter selection. J Stat Plan Inference 137(9):2784–2801
Ramsay JO, Silverman BW (2002) Applied functional data analysis: methods and case studies. Springer, New York
Ramsay JO, Silverman BW (2005) Functional data analysis. Springer, New York
Sarda P, Vieu P (2000) Kernel regression. In: Schimek M (ed) Smoothing and regression, approaches, computation and application. Wiley, New York, pp 43–70
Vilar-Fernández JM, Cao R (2007) Nonparametric forecasting in time series. A comparative study. Commun Stat Simul Comput 36:311–334
Zeidler E (1986) Nonlinear functional analysis and its applications I: fixed-point theorems. Springer, New York
Acknowledgements
The authors of this article would like to express their gratitude to the three anonymous reviewers for their constructive critiques, suggestions, and comments that have led to improved version of the article. The authors are also grateful to the Editor-in-Chief for patiently handling the article.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Proof of Lemma 1
Note that
The second equality follows from the change of variable \(t=d/h\). Noticing the definition of \(\tau _h(s)\) and \(\tau _h(1)=1\) and using integration by parts we then have
Thus
This completes the proof of Lemma 1. \(\square \)
Proof of Theorem 1
Recall that for random variables U and V the following identity holds
from which with \(U={{\widehat{r}}}_{\scriptscriptstyle h,2}(x)\) and \(V={{\widehat{r}}}_{\scriptscriptstyle h,1}(x)\), the following classical decomposition of the bias term \(B_n\) of the estimator \({{\widehat{r}}}_{\scriptscriptstyle h}(x)\) follows
where \({B_{n,1}}={I\!\!E}\left[ {{\widehat{r}}}_{\scriptscriptstyle h,1}(x))({{\widehat{r}}}_{\scriptscriptstyle h,2}(x)-{I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}(x))\right] \) and \({B_{n,2}}={I\!\!E}\Big [{{\widehat{r}}}_{\scriptscriptstyle h}(x)\Big ({{\widehat{r}}}_{\scriptscriptstyle h,1}(x)-{I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,1}(x))\Big )^2\Big ]\).
From the independence assumption of X and \(\varepsilon \) we have
Write \(\omega _i(x)= \sum ^{n}_{j=1} \omega _{j,i}(x)\), where
with \(\omega _{i,i}(x)=0\). Then \({I\!\!E}(\omega _i(x))= \sum ^{n}_{j=1}{I\!\!E}(\omega _{j,i}(x))\). Using the assumption that \(X_i's\) are independent as in hypothesis (8), and denoting \(d_j = d(X_j,x)\) to save space, we have
From Lemma 1, with \(k=1, 2\), and \(\nu =1\), respectively, we have
Therefore, asymptotically, the denominator D becomes
For the numerator term N, we notice that
Using Taylor expansion of \(\phi \) at zero (noting \(\phi (0)=0\)) up to order two and hypotheses (3) and (7), we have
and similarly,
which together with Lemma 1 yield
and thus the numerator N has the following asymptotic expression
Therefore
The rate of convergence of the term \(B_{n,1}\) follows from assertion (iii) of Lemma 3:
For the term \(B_{n,2}\), by Cauchy-Shwartz inequality, we have
and using (i) and (ii) of Lemma 3, it follows that
Therefore, \(B_{n,1}\) and \(B_{n,2}\) are asymptotically negligible with respect to \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}(x))/{I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,1}(x))\). \(\square \)
Proof of Theorem 2
We use the following decomposition of the variance as in, for instance, Sarda and Vieu (2000) in the finite dimensional case and by Ferraty et al. (2007) in the functional framework.
The asymptotic expression of the variance is obtained from Lemmas 2 and 3:
This completes the proof of Theorem 2. It remains to prove Lemmas 2 and 3. \(\square \)
Proof of Lemma 2
The asymptotic expression of
follows from the proof of Theorem 1. Concentrating on \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}(x))\), we have
where \(\omega _{j,i}(x)\) is defined in (10), and since the \(X_i's\) are assumed independent, we have
From (11) in the proof of Theorem 1, we have
so that
and hence,
\(\square \)
Proof of Lemma 3
Proof of (i). From the following definition of the variance
and the expression of \({{\widehat{r}}}_{\scriptscriptstyle h,1}(x)\), we can write
Now,
Likewise from hypothesis (8), we have
From Lemma 1, we obtain
Now, we compute \({I\!\!E}(\omega _{j_1,i}(x)\omega _{j_2,i}(x))\) for \(j_1 \ne j_2 \ne i\):
Again, using Lemma 1, we obtain
It follows that
where \(\psi _1= M_{0,2}M_{4,0}-2M_{1,2}M_{3,1}+ M^2_{2,2}\) and \(\psi _2\) is defined in Theorem 2.
Concentrating on the second term of \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,1}^2(x))\), we have for \(i_1 \ne i_2\)
We have for \(j_1 \ne i_1\),
Then from Lemma 1, we obtain
where \(\psi _3\) is defined in Theorem 2.
For \(j_1 \ne j_2\) (\(\ne i_1,i_2\)),
so that from Lemma 1,
where \(\psi _4= M^2_{2,1}M^2_{0,1} -2M_{2,1}M^2_{1,1}M_{0,1} +M^4_{1,1}\). Therefore, we obtain
Gathering the terms (13) and (14), we obtain
and noticing that \(\psi _4 = \psi ^2_0\), we finally obtain the asymptotic expression of the variance:
Proof of (ii). Using the following definition of the variance
and the expression of \({{\widehat{r}}}_{\scriptscriptstyle h,2}(x)\), we write
Now, for the first term of \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}^2(x))\), we have
and with (13), we obtain
For the second term of \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}^2(x))\), for \(i_1\ne i_2\), since \(X_i's\) and \(\epsilon _i's\) are independent, we have
so that from (14), we have
Collecting the above two terms, we obtain
Since \(n\varphi (h)\rightarrow \infty \) and \(\psi _4=\psi ^2_0\), it follows that
This completes the proof of assertion (ii).
Proof of (iii). We write
with
The same techniques as in the proof of (ii) are applied to obtain
and
It follows that
Therefore using the asymptotic expressions of \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,1}(x)\) and \({I\!\!E}({{\widehat{r}}}_{\scriptscriptstyle h,2}(x)\) as in assertion (i) and (15), we obtain
This completes the proof of Lemma 3. \(\square \)
Proof of Corollary 2
In Corollary 1, we replace the autocovariance function \(\rho _\epsilon (j)\) by \(\mathcal {C}\,|j|^{-\gamma }\) and using that \( \sum _{j=1}^n j^{-\gamma }= O(n^{1-\gamma })\), we have \(\frac{{\tilde{s}}_n}{n^2}\approx \frac{s_n}{n}= O(n^{-\gamma })\), where \(s_n=\sum ^{n}_{k=1}\left| \rho _\epsilon (k)\right| \). We deduce
and if \(n^{1-\gamma } \varphi (h) \rightarrow \infty \), then
This completes the proof of Corollary 2. \(\square \)
Proof of Propositions 1 and 2
It suffices to choose the bandwidth that minimizes the asymptotic MSE of Corollary 1
and to use the expression of the small probability \(\varphi (h)\sim Ch^\tau \).
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Benhenni, K., Hajj Hassan, A. & Su, Y. The effect of correlated errors on the performance of local linear estimation of regression function based on random functional design. Stat Papers (2024). https://doi.org/10.1007/s00362-023-01523-z
Received:
Revised:
Published:
DOI: https://doi.org/10.1007/s00362-023-01523-z