Abstract

This paper proposes a novel Function-on-Function (FoF) quantile re

gression model for characterizing the influence of functional covariates across

different parts of the conditional distribution of functional responses. We design

a loss function based on geometric quantiles in infinite-dimensional spaces. This

ensures that our method provides well-defined conditional quantiles for functional

data, while retaining desirable statistical properties similar to those in the univariate setting. Additionally, the convergence rates of estimators are explicitly

derived for both given and estimated basis functions. To address the challenges

posed by large-scale functional data that are often distributed across multiple

computing nodes, we further extend our framework to a distributed setting. We

propose a communication-efficient smooth FoF geometric quantile regression algorithm that not only achieves computational and communication efficiency but

also guarantees convergence rates comparable to those of global estimators.

Key words and phrases: Communication efficiency, Distributed computation, Functional data, Function-on-function regression, Geometric quantile

Information

Preprint No.SS-2026-0148
Manuscript IDSS-2026-0148
Complete AuthorsBingchen Li, Alan T.K. Wan, Miaomiao Yu, Yong Zhou
Corresponding AuthorsMiaomiao Yu
Emailsmmyu@fem.ecnu.edu.cn

References

  1. Benatia, D., M. Carrasco, and J.-P. Florens (2017). Functional linear regression with functional response. Journal of Econometrics 201(2), 269–291.
  2. Beyaztas, U., H. L. Shang, and A. Alin (2022). Function-on-function partial quantile regression. Journal of Agricultural, Biological and Environmental Statistics 27, 149–174.
  3. Beyaztas, U., H. L. Shang, and S. Saricam (2025). Penalized function-on-function linear quantile regression. Computational Statistics 40, 1–29.
  4. Bosq, D. (2000). Linear processes in function spaces: theory and applications, Volume 149. Springer Science & Business Media.
  5. Cardot, H., P. C´enac, and P.-A. Zitt (2013). Efficient and fast estimation of the geometric median in Hilbert spaces with an averaged stochastic gradient algorithm. Bernoulli 19(1), 18–43.
  6. Chakraborty, A. and P. Chaudhuri (2014). The spatial distribution in infinite dimensional spaces and related quantiles and depths. Annals of Statistics 42, 1203–1231.
  7. Chen, X., J. D. Lee, H. Li, and Y. Yang (2022). Distributed estimation for principal component analysis: An enlarged eigenspace analysis. Journal of the American Statistical Association 117(540), 1775–1786.
  8. Chowdhury, J. and P. Chaudhuri (2019). Nonparametric depth and quantile regression for functional data. Bernoulli 25, 395–423.
  9. Dong, C., R. Chen, Z. Xiao, and W. Liu (2024). Functional quantile autoregression. Journal of Econometrics 244(2), 105765.
  10. Fan, J., D. Wang, K. Wang, and Z. Zhu (2019). Distributed estimation of principal eigenspaces. The Annals of Statistics 47(6), 3009 – 3031.
  11. He, G., H. G. M¨uller, J. L. Wang, and W. Yang (2010). Functional linear regression via canonical analysis. Bernoulli 16, 705–729.
  12. Horv´ath, L., P. Kokoszka, and S. Wang (2021). Monitoring for a change point in a sequence of distributions. Annals of Statistics 49, 2271–2291.
  13. Hu, B., X. Hu, H. Liu, J. You, and J. Cao (2024). Simultaneous functional quantile regression. Statistica Sinica 34(2), 867–888.
  14. Jordan, M. I., J. D. Lee, and Y. Yang (2019). Communication-efficient distributed statistical inference. Journal of the American Statistical Association 114(526), 668–681.
  15. Kato, K. (2012). Estimation in functional linear quantile regression. Annals of Statistics 40, 3108–3136.
  16. Konen, D. (2025). PDE characterization of geometric distribution functions and quantiles. Bernoulli 31, 2077–2104.
  17. Li, B., A. T. K. Wan, M. Yu, and Y. Zhou (2026). Supplement to “Function-on-Function Geometric Quantile Regression with a Distributed Extension”.
  18. Li, C., L. Xiao, and S. Luo (2022). Joint model for survival and multivariate sparse functional data with application to a study of Alzheimer’s Disease. Biometrics 78, 435–447.
  19. Li, Y., K. Chen, X. Zheng, and C. Y. Yau (2024). Functional threshold autoregressive model. Statistica Sinica 34(2), 817–836.
  20. Liu, Y., M. Li, and J. S. Morris (2020). Function-on-scalar quantile regression with application to mass spectrometry proteomics data. Annals of Applied Statistics 14(2), 521.
  21. Ma, H., T. Li, H. Zhu, and Z. Zhu (2019). Quantile regression for functional partially linear model in ultra-high dimensions. Computational Statistics & Data Analysis 129, 135–147.
  22. Mutis, M., U. Beyaztas, F. Karaman, and H. L. Shang (2025). On function-on-function linear quantile regression. Journal of Applied Statistics 52, 814–840.
  23. Shen, S., J. Lu, and X. Lin (2026). Dimension reduction for large-scale federated data: Statistical rate and asymptotic inference. Journal of the American Statistical Association 121(553), 585–597.
  24. Yan, X. and S. He (2026). Poisson subsampling for large-scale functional data analysis. Computational Statistics 41, 101.
  25. Yao, F., H. G. M¨uller, and J. L. Wang (2005). Functional linear regression analysis for longitudinal data. Annals of Statistics 33, 2873–2903.
  26. Zhang, Z., X. Wang, L. Kong, and H. Zhu (2022). High-dimensional spatial quantile functionon-scalar regression. Journal of the American Statistical Association 117(539), 1563–1578.
  27. Zhu, H., Y. Zhang, Y. Li, and H. Lian (2023). Semiparametric function-on-function quantile regression model with dynamic single-index interactions. Computational Statistics and Data Analysis 182, 107727. Bingchen Li, Miaomiao Yu and Yong Zhou

Acknowledgments

We thank the anonymous referees, the Associate Editor, and the Editor

for their constructive comments, which have helped improve the quality of

this paper. This work is supported in part by the National Natural Science

Foundation of China (72301108), the State Key Program of the National

Natural Science Foundation of China (72531003 and 72331005), the National Key R&D Program of China (2021YFA1000100 and 2021YFA1000101),

the Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the

Ministry of Education of China (JYB2025XDXM904), the Shanghai Pilot

Program for Basic Research (TQ20240201), the Natural Science Foundation of Shanghai (23JS1400500), and the Shanghai Municipal Education

Commission (2024AI01002).

Supplementary Materials

The Supplementary Material contains preliminaries, detailed algorithms, a

discussion of the differences between DFPCA and existing distributed PCA

methods, implementation details for the simulation studies, discussions of

link function misspecification and nonparametric link modeling, operator

convergence rates, additional simulation results, further details of the real

case study, and complete proofs of the theoretical results.