Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Spatial function-on-function regression

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read The paper claims that the infinite-dimensional spatial function-on-function regression model can be rewritten exactly as a finite multivariate spatial autoregressive model, and that least-squares estimates of the coefficient functions are…

desk verdict Genuinely new combination for spatial function-on-function regression, but the asymptotics only cover an exact finite-rank case the simulations don't satisfy; fixable with a truncation-bound argument. read the letter →

arxiv 2412.17327 v1 pith:HQHKSDGU submitted 2024-12-23 stat.ME stat.CO

classification stat.MEstat.CO MSC 62R1062M3062J0562G20
keywords function-on-functionregressionspatialautoregressivemodelfunctionalprincipalcomponentanalysisdependencemultivariateleastsquaresestimationdataCOVID-19casestudy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper introduces a spatial function-on-function regression (SFoFR) model in which the functional response at each location depends on the functional responses of its neighbors through a spatial lag integral, in addition to the usual functional covariate effect. The central claim is that projecting the response onto spatial functional principal components and the predictor onto ordinary functional principal components converts the infinite-dimensional model exactly into the finite-dimensional multivariate spatial autoregressive equation $Y^{\top} = WY^{\top}\rho + X^{\top}\beta + e^{\top}$. From that reduction, the paper obtains least-squares estimators of the spatial autocorrelation function $\rho(u,t)$ and the regression coefficient function $\beta(s,t)$, and proves they are $\sqrt{n}$-consistent and asymptotically normal. Simulations show the estimator outperforms functional partial least squares and functional PCA when spatial dependence is moderate to strong, and an application to Brazilian COVID-19 data reports improved in-sample and out-of-sample prediction.

What carries the argument

The machinery is principal-component projection followed by algebraic collapse: spatial functional principal component (SFPC) analysis on the response and classical functional principal component (FPC) analysis on the predictor rewrite each infinite-dimensional object using a finite score vector, and orthonormality of the eigenfunctions yields the identity $Y^{\top} = WY^{\top}\rho + X^{\top}\beta + e^{\top}$ (equation 3.12). This identity turns the original inverse problem into a multivariate spatial autoregressive model whose coefficient matrices $\rho$ and $\beta$ are estimated by least squares; multiplying the estimated scores back by the eigenfunctions gives the functional estimates $\hat{\rho}(u,t)$ and $\hat{\beta}(s,t)$.

What would settle it

One concrete check: in the paper's strong-dependence simulation setting, $\rho(u,t)=0.9(1+ut)/(1+|u-t|)$ gives $\|\rho\|_\infty = 1.8$ for a row-normalized $W$ with $\|W\|_\infty = 1$, violating the stated contraction condition $\|\rho\|_\infty < 1/\|W\|_\infty$ from Proposition 2.1, yet the method still works — so computing the spectral radius of the corresponding spatial operator would show whether the true stability condition is much weaker than the paper's proposition claims.

Watch

Extended reading notes

Core claim

The paper establishes that estimating the two functional parameters in the SFoFR model is algebraically equivalent to estimating a finite multivariate spatial autoregressive (MSAR) model once the data are expressed in principal component coordinates. Substituting the truncated Karhunen-Loève expansions of $Y(t)$, $X(s)$, $\rho(u,t)$, and $\beta(s,t)$ into equation (2.3) and using the orthonormality of the basis functions collapses the infinite-dimensional model to $Y^{\top} = WY^{\top}\rho + X^{\top}\beta + e^{\top}$, with no loss beyond the truncation itself. The paper then imports the least-squares estimator of Zhu et al. (2020) for the vectorized MSAR equation and, under regularity conditions that include the true functions lying in the span of the retained eigenfunctions, proves the recovered functional estimates are $\sqrt{n}$-consistent and converge weakly to a Gaussian process.

Load-bearing premise

The theoretical claims require the true spatial autocorrelation and regression coefficient functions to lie exactly in the span of the first estimated principal component functions; if that fails, truncation bias makes the estimator consistent for an approximation rather than the truth.

Editorial extensions

If this is right

  • Because the infinite-dimensional model collapses to a finite MSAR equation, estimation reduces to a least-squares computation on score matrices, making the method computationally feasible for large areal datasets.
  • The $\sqrt{n}$-consistency and asymptotic normality results give a covariance operator from which confidence bands for $\rho(u,t)$ and $\beta(s,t)$ can be constructed, enabling inference about the strength of spatial dependence.
  • In the paper's simulations, SFoFR attains lower integrated squared error for $\beta(s,t)$ than functional partial least squares and functional PCA when spatial dependence is moderate to strong ($\alpha=0.5$ and $\alpha=0.9$), while the comparators degrade substantially.
  • On the Brazilian COVID-19 data, the SFoFR model yields higher $R^2$ and $R^2_{\text{new}}$ values than FPLS and FPC, and its estimated spatial autocorrelation function tracks the functional Moran's I pattern across the two years.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reduction is purely algebraic, so the same projection-to-scores strategy should extend to spatial error models, spatial Durbin forms, multiple functional predictors, and scalar covariates, provided the corresponding multivariate spatial estimator exists.
  • The sup-norm contraction condition in Proposition 2.1 is sufficient but likely not necessary; in the paper's own $\alpha=0.9$ setting the condition is violated, yet the Neumann series simulations converge, suggesting the operative requirement is a spectral-radius condition on the spatial operator.
  • A direct finite-sample check of the theorem would be to simulate under condition C3, construct Gaussian-process confidence bands using the derived covariance operator, and measure coverage; any substantial undercoverage would point to a missing term in the covariance formula.
  • If users choose $K$ too small, the estimator will converge to the projection of the truth rather than the truth itself; comparing estimates across increasing $K$ offers a practical diagnostic for truncation bias.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes a spatial function-on-function regression (SFoFR) model, given in (2.2)-(2.3), in which the functional response at each areal unit depends on a spatial lag of the response through a bivariate kernel rho(u,t) and on a functional covariate through beta(s,t). The estimation strategy projects the response onto spatial functional principal components (SFPC) and the predictor onto classical functional principal components, rewrites the infinite-dimensional model as the multivariate spatial autoregressive model (3.12), and applies the least-squares estimator of Zhu et al. (2020) to the score vectors. The authors state sqrt(n)-consistency and asymptotic normality for the estimated functions (Theorem 3.1), support the method with simulations against FPC and FPLS competitors, and illustrate it on Brazilian COVID-19 data, with an R package available.

Significance. The algebraic reduction from model (2.3) to the MSAR form (3.12) is clean, and, once the exact finite-rank representation is accepted, the use of the Zhu et al. (2020) estimator is natural and computationally attractive. The paper ships an R package and conducts a genuinely out-of-sample COVID-19 evaluation (2021 training, 2022 prediction), which are strengths that deserve explicit credit. The substantive contribution, adding a SAR-type spatial dependence component to function-on-function regression, fills a real gap. However, the advertised operating regime, an infinite-dimensional model, is not what Theorem 3.1 proves: the theorem holds only under exact finite-rank conditions that the paper's own simulation design does not satisfy, and no truncation-bias analysis is supplied. The contribution is therefore conditional on a nontrivial extension.

major comments (3)
  1. [Section 3.1 and supplement, Condition C3] The exact equality (3.12) and Theorem 3.1 require beta(s,t) and rho(u,t) to lie exactly in the span of the first Kx and Ky eigenfunctions. In the simulation study (Section 4), beta(s,t)=2+s+t+0.5 sin(2*pi*s*t) and rho(u,t)=alpha(1+ut)/(1+|u-t|) are smooth infinite-rank functions, and Kx and Ky are chosen by a 95% cumulative variance rule rather than by exact representation. Hence the reported estimators are least-squares estimates of a truncated or projected model, and Theorem 3.1 does not cover the reported finite-sample results. The paper must either supply a growing-K truncation-bias bound that establishes consistency for the true functions or redesign the simulations so that the data-generating parameters satisfy C3.
  2. [Section 4, simulation design vs. Proposition 2.1] The text states that the response is generated with a Neumann series and that rho(u,t) satisfies ||rho||_inf < 1/||W||_inf, as required by Proposition 2.1. For alpha=0.9, ||rho||_inf = max_{u,t in [0,1]^2} 0.9(1+ut)/(1+|u-t|) = 1.8, while the row-normalized weight matrix has ||W||_inf = 1, so the stated sufficient condition is violated. The generated process may still be stable because the condition is only sufficient, but the justification written in the paper is incorrect; the authors should verify stability directly, for example through the spectral radius of the induced integral operator, or choose a rho that satisfies the stated condition.
  3. [Supplement, proof of Theorem 3.1] The proof imports Lemmas 2-6 from Zhu et al. (2020) and Condition C9 directly, but it does not verify the hypotheses of those lemmas when the scores are obtained from estimated eigenfunctions and when Kx and Ky are either fixed or growing. In particular, the asymptotic covariance operator is expressed in terms of the estimated basis, yet no argument shows that the uniform convergence in C1 is sufficient to control the effect of basis estimation inside the least-squares objective. This gap must be closed before the main asymptotic claim can be accepted in the functional setting.
minor comments (6)
  1. [Section 4, discussion of Tables 1 and 2] The weak-spatial-dependence cases are labeled 'alpha = 1' in the text, but the simulations use alpha=0.1; the labels should be corrected.
  2. [Section 3.1, equations (3.10)-(3.13)] The symbol Y^T denotes the n x Ky matrix of scores rather than the transpose of the vector-valued curve Y(t); a distinct symbol such as Y_scores would avoid confusion, especially because eY=vec(Y^T) is introduced immediately afterward.
  3. [Abstract] The sentence 'The [package anonymized for review] package in provides...' is incomplete because of anonymization; the package name and language should be restored in the published version.
  4. [Section 1, Introduction] The phrase 'under some regulatory conditions' should read 'regularity conditions.'
  5. [Supplement, proof of Proposition 2.1] The bound is attributed to Jensen's inequality, but the step is an application of the Minkowski integral inequality followed by ||Y||_1 <= ||Y||_p on [0,1]; the result is correct, but the justification should be corrected.
  6. [Figure 1 caption] The caption says '100 generated sample curves,' but the simulations use n_train in {100, 250, 500, 1000}; the displayed sample size should be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the SFoFR-to-MSAR reduction is a basis-expansion identity, identification and asymptotics come from an external source, and the empirical evaluations are genuinely out-of-sample.

full rationale

The derivation chain is not circular. Equation (3.12) is obtained by substituting the basis expansions (3.5)-(3.9) into (2.3), using orthonormality of phi and psi, and integrating out t; this is an algebraic projection identity rather than a definition of the target in terms of the estimator. The estimators b_rho and b_beta in (3.16)-(3.17) are then least-squares solutions for the projected MSAR coefficients, with identification and sqrt(n)-asymptotic normality imported from Zhu et al. (2020), an external source; SFPC is taken from Khoo et al. (2023), also external. No fitted parameter is relabeled as a prediction: the COVID-19 analysis trains on 2021 and evaluates on 2022, and the simulations generate independent test sets from the same design. The paper's own earlier FoFR work appears only as background citation and is not load-bearing. The main caveats, namely that Theorem 3.1 assumes exact finite-rank representation (Condition C3) while the simulation functions are not finite-rank combinations of the estimated empirical eigenbases, and that the alpha=0.9 design actually violates the sufficient condition of Proposition 2.1, are matters of assumption coverage and proof applicability rather than circularity; they do not make any result equal to its input by construction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The contribution rests on imported tools (SFPC, MSAR least squares) and the paper's own assumptions about exact truncation and contraction. The free parameters are truncation ranks, B-spline basis sizes, and the KNN window h; none are reported with values, which weakens reproducibility.

free parameters (3)
  • Kx, Ky truncation ranks = chosen by 95% variance criterion; values not reported
    The number of FPC/SFPC components determines the MSAR dimensions and therefore the estimates; results depend on this threshold.
  • B-spline basis sizes Lx, Ly = not stated
    All curves are expanded in B-splines before eigenanalysis, but the paper never gives the number or placement of knots, affecting numerical estimates.
  • Number of nearest neighbors h in COVID-19 application = not reported
    W is constructed from h-nearest neighbors with h selected by cross-validation; the selected h is omitted, so the weight matrix cannot be reproduced.
assumptions (5)
  • standard math Karhunen-Loeve expansion with distinct eigenvalues and square-integrable kernels
    Used in Section 3 to represent Y, X, rho, beta, and epsilon by eigenfunctions.
  • domain assumption The spatial weight matrix W is fixed, known, and row-normalized
    Section 2 defines W from distances; estimation and theory condition on W, and misspecification of W is not studied.
  • ad hoc to paper Condition C3: beta(s,t) and rho(u,t) lie exactly in the span of the first Kx/Ky eigenfunctions
    Makes the reduction to equation (3.12) exact; no truncation error term is analyzed.
  • ad hoc to paper Proposition 2.1 contraction bound ||rho||_inf < 1/||W||_inf
    Guarantees existence, uniqueness, and Neumann series convergence; violated for alpha=0.9 in the simulations.
  • standard math Zhu et al. (2020) Lemmas 1-6 for MSAR least squares
    Theorem 3.1 imports these external results for invertibility, identification, and asymptotic normality.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Spatial function-on-function regression." pith.science (2026). https://pith.science/paper/HQHKSDGU

@misc{pith2026241217327,
  author       = {Pith},
  title        = {Pith review of: Spatial function-on-function regression},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HQHKSDGU}},
  note         = {Machine review of arXiv:2412.17327}
}
read the original abstract

We introduce a spatial function-on-function regression model to capture spatial dependencies in functional data by integrating spatial autoregressive techniques with functional principal component analysis. The proposed model addresses a critical gap in functional regression by enabling the analysis of functional responses influenced by spatially correlated functional predictors, a common scenario in fields such as environmental sciences, epidemiology, and socio-economic studies. The model employs a spatial functional principal component decomposition on the response and a classical functional principal component decomposition on the predictor, transforming the functional data into a finite-dimensional multivariate spatial autoregressive framework. This transformation allows efficient estimation and robust handling of spatial dependencies through least squares methods. In a series of extensive simulations, the proposed model consistently demonstrated superior performance in estimating both spatial autocorrelation and regression coefficient functions compared to some favorably existing traditional approaches, particularly under moderate to strong spatial effects. Application of the proposed model to Brazilian COVID-19 data further underscored its practical utility, revealing critical spatial patterns in confirmed cases and death rates that align with known geographic and social interactions. An R package provides a comprehensive implementation of the proposed estimation method, offering a user-friendly and efficient tool for researchers and practitioners to apply the methodology in real-world scenarios.

Figures

Figures reproduced from arXiv: 2412.17327 by the authors.

Figure 1
Figure 1. Graphs depicting 100 generated sample curves for functional response (top left panel) and functional covariate (top right panel), along with true bivariate regression coefficient functions (bottom left panel) and true spatial autocorrelation parameter function (bottom right panel). The data are generated with exponential distance weight matrix and when α = 0.9. then evaluated on these test samples. Specifically, we … view at source ↗
Figure 2
Figure 2. Choropleth maps depicting the average confirmed cases (left panels) and the average number of deaths (right panels) for Brazilian cities in the years 2021 (top panels) and 2022 (bottom panels). In our spatial correlation analysis of the Brazilian COVID-19 dataset, we compute the functional Moran’s I statistic to assess the degree of spatial autocorrelation. To begin, we construct the spatial weight matrix W = (wii′)… view at source ↗
Figure 3
Figure 3. The graphical display of Brazil’s COVID-19 data presents the confirmed cases in the left panels and the average number of deaths in the right panels for 2021 (top panels) and 2022 (bottom panels). Different colors represent different cities, with the observations of confirmed cases being functions of days, i.e., 1 ≤ t ≤ 365. KNN-based weight matrix is then defined as: wii′ =    Nh(i) −1 , if i ′ ∈ Nh(i) 0, ot… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Functional Moran’s I statistics for 2021 (left panel) and 2022 (right panel) were calculated using the KNN connection network [PITH_FULL_IMAGE:figures/full_fig_p030_4.png]
Figure 5
Figure 5. Figure 5: Estimated bivariate regression parameter functions for the Brazilian COVID-19 dataset obtained by the methods FPLS (left panel), FPC (middle panel), and SFoFR (right panel). The estimated spatial autocorrelation parameter function, obtained using the proposed method, i…
Figure 6
Figure 6. Figure 6: Estimated spatial autocorrelation parameter function for the Brazilian COVID-19 dataset by the proposed SFoFR. 6 Conclusion We introduced a novel SFoFR model to address complex spatial dependencies inherent in functional data. By combining spatial autoregressive modeli…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Spatial Functional Deep Neural Network Model: A New Prediction Algorithm

    stat.ME 2025-04 reject novelty 4.0 of 10

    A hybrid model that estimates spatial dependence by maximum likelihood and then feeds spatially adjusted functional and scalar inputs into a deep neural network predicts spatially dependent outcomes with lower test er...

Reference graph

Works this paper leans on

7 extracted references · 6 canonical work pages · cited by 1 Pith paper

  1. [1]

    Acal, C., Escabias, M., Aguilera, A. M. and Valderrama, M. J. (2021), ‘COVID-19 data imputation by multiple function-on-function principal component regression’, Mathematics 9(11),

  2. [8]

    and Gotway, C

    Schabenberger, O. and Gotway, C. A. (2017),Statistical Methods for Spatial Data Analysis, Chapman and Hall/CRC, New York. Scheipl, F., Staicu, A.-M. and Greven, S. (2015), ‘Functional additive mixed models’, Journal of Compu- tational and Graphical Statistics 24(2), 477–501. Wang, B. and Xu, A. (2019), ‘Gaussian process methods for nonparametric functiona...

  3. [130]

    A., Gayawan, E., Elfaki, F., Kuddus, M.-A

    Saffary, T., Adegboye, O. A., Gayawan, E., Elfaki, F., Kuddus, M.-A. and Saffary, R. (2020), ‘Analysis of COVID-19 cases’ spatial dependence in US counties reveals health inequalities’, Frontiers in Public Health

  4. [324]

    (1995), ‘Local indicators of spatial association–LISA’, Geographical Analysis 27(2), 93–115

    Anselin, L. (1995), ‘Local indicators of spatial association–LISA’, Geographical Analysis 27(2), 93–115. Anselin, L. (1998), Spatial Econometrics: Methods and Models, Springer, Berlin. Beyaztas, U. and Shang, H. L. (2020), ‘On function-on-function regression: Partial least squares approach’, Environmental and Ecological Statistics 27(1), 95–114. Beyaztas,...

  5. [674]

    S., Staicu, A.-M., Maity, A., Carroll, R

    Kim, J. S., Staicu, A.-M., Maity, A., Carroll, R. J. and Ruppert, D. (2018), ‘Additive function-on-function regression’, Journal of Computational and Graphical Statistics 27(1), 234–244. Kirk, W. A. and Khamsi, M. A. (2001), An Introduction to Metric Spaces and Fixed Point Theory, John Wiley, New York. Lesage, J. and Pace, R. K. (2009), Introduction to Sp...

  6. [1237]

    C., Durban, M

    Aguilera-Morillo, M. C., Durban, M. and Aguilera, A. M. (2017), ‘Prediction of functional data with spatial dependence: A penalized approach’, Stochastic Environmental Research and Risk Assessment 31(1), 7–22. Almalki, A., Gokaraju, B., Acquaah, Y. and Turlapaty, A. (2022), ‘Regression analysis for COVID-19 infections and deaths based on food access and h...

  7. [2376]

    (2024), Functional spatial autoregressive models, Technical report, arXiv

    40 Hoshino, T. (2024), Functional spatial autoregressive models, Technical report, arXiv. URL: https: //arxiv.org/pdf/2402.14763. Huang, T., Saporta, G., Wang, H. and Wang, S. (2021), ‘A robust spatial autoregressive scalar-on- function regression with t-distribution’, Advances in Data Analysis and Classification 15(470), 57–81. Ivanescu, A. E., Staicu, A...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.