Simultaneous Sieve Estimation and Inference for Time-Varying Nonlinear Time Series Regression

Xiucai Ding; Zhou Zhou

REVIEW 3 major objections 5 minor 88 references

Simultaneous Sieve Estimation and Inference for Time-Varying Nonlinear Time Series Regression

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Mapped sieve bases let time-varying nonlinear regression be estimated and covered uniformly over [0,1]×R, with bootstrap-based simultaneous confidence regions.

desk verdict A serious, well-built framework for simultaneous inference in time-varying nonlinear regression, but the unbounded-domain approximation result has a genuine gap: the mapped sieve space only contains functions vanishing at infinity, and Assumption 3.1 doesn't require that. read the letter →

arxiv 2506.23069 v1 pith:BVRUSFMA submitted 2025-06-29 stat.ME math.STstat.TH

Xiucai Ding , Zhou Zhou This is my paper · ORCID

classification stat.MEmath.STstat.TH MSC 62M1062G0862G0562G10

keywords nonstationarytimeseriestime-inhomogeneousnonlinearregressionsievemethodsimultaneousinferenceGaussianapproximationmultiplierbootstraplocallystationaryprocessesmappedbasisfunctions

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

Time-varying nonlinear regression with locally stationary covariates is hard because the regression functions live on [0,1]×R, so classical sieve approximations on a compact box fail near the tails. This paper claims a mapped, hierarchical sieve expansion approximates such functions uniformly over the whole domain, and that a two-step OLS correction makes the resulting estimators uniformly consistent while restoring the centering condition that identifies each additive component. On top of that, the paper builds simultaneous confidence regions covering every time point and covariate value at asymptotic level 1−α, with a multiplier bootstrap that produces the critical values. If correct, this turns sieve estimation into a tool for structural testing—exact form, time-homogeneity, separability—without kernel boundary corrections or compact-support restrictions.

What carries the argument

The paper's workhorse is the mapped hierarchical sieve basis: a monotone map g(y;s) sends the unbounded covariate domain (R or R+) to [−1,1], and tensor products of an orthonormal basis in t with the mapped basis in x form the approximation space (3.8). Because the mapping is smooth, smoothness of m_j(t,x) in x becomes smoothness of em_j(t,y) on a compact square, so classical sieve approximation rates transfer to unbounded domains. A second OLS step estimates the time-varying mean shift χ_j(t)=E[m_{j,c,d}(t,X_{j,i})] and subtracts it from the pilot estimator, which enforces the identifiability condition (1.2); Gaussian approximation and volume-of-tubes devices then control the sup-norm of the estimation error and the critical value.

What would settle it

Take a model in which one regression function does not decay as |x|→∞—for example m_j(t,x)=cos(x) or m_j(t,x)=2+sin(x)—so the mapped function is not smooth at the boundary of [0,1]. Then the claimed uniform approximation rate O($c^{{−m1j}}$+$d^{{−m2j}}$) should fail; a simulation could check whether the sup-norm error of the sieve estimator over a growing covariate interval [−L,L] fails to shrink as L grows, or whether the nominal 95% simultaneous coverage drops well below 95%.

Watch

Extended reading notes

Core claim

The central claim is that for model (1.1), where each m_j is smooth and decays sufficiently fast as |x|→∞, the bias-corrected sieve estimators (3.21)-(3.22) are uniformly consistent over [0,1]×R, and the simultaneous confidence regions (4.7)/(4.28) have asymptotic coverage exactly 1−α. The proof rests on three technical pillars: a uniform approximation theorem (Proposition 3.1) for 2-D functions on unbounded domains via mapped sieve bases; two Gaussian approximation results for affine forms of high-dimensional locally stationary time series (Theorems L.2 and L.3); and a volume-of-tubes expansion (Theorem 4.2) for the critical value of the maximum of the resulting Gaussian field. The multiplier bootstrap (Theorem 4.3) makes the construction operational by approximating both the variance function h_j(t,x) and the critical value from one realization.

Load-bearing premise

The foundation is Assumption 3.1: after mapping the covariate domain to [0,1], each regression function must be smooth with uniformly bounded derivatives, which in practice means the original functions decay rapidly as |x|→∞; if a true function does not decay at infinity, the approximation rate in Proposition 3.1 and everything built on it collapses.

Editorial extensions

If this is right

If the central claim holds, practitioners can construct simultaneous 1−α confidence regions for each time-varying regression function over the entire unbounded covariate range from a single observed time series.
Structural tests for time-invariance, multiplicative separability, and exact parametric form are valid at asymptotic level α and have power tending to 1 for deviations larger than the order of the band width.
The estimator achieves the optimal uniform rate O(n^{−1/2} log^3 n) when the regression functions are infinitely smooth and the sieve dimensions grow logarithmically with n.
The multiplier bootstrap procedure is theoretically sound and is implemented in an accompanying R package, so the method is ready for routine use.
The two Gaussian approximation results for affine forms of high-dimensional locally stationary time series are stated as having independent interest beyond this regression setting.

Reading between the lines

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

The same mapped-sieve and Gaussian-approximation machinery likely transfers to locally stationary nonlinear autoregressions with lagged covariates, which the paper notes requires only minor changes; this is our inference, not a theorem of the paper.
Because the approximation rate depends on how fast the mapped functions approach the boundary, choosing the map's scale parameter data-adaptively, rather than fixing s=1 by convention, could noticeably improve finite-sample coverage.
The volume-of-tubes formula is tailored to a Gaussian field indexed by a 2-D manifold; for higher-dimensional covariate vectors the same reasoning would require a higher-dimensional manifold and a different critical-value expansion, an extension the paper does not pursue.
The identifiability correction is estimated rather than imposed by design, which suggests the method may also work when covariates are mutually dependent, a setting where standard additive-model centering is harder to justify.

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.

Desk Editor's Note

read the letter

You should know two things about this paper. First, it is a serious and unusually complete contribution: it claims the first simultaneous inferential method for the general model (1.1), and the claim holds up against the cited literature. The proof architecture is coherent — uniform consistency, pointwise and uniform Gaussian approximation, volume-of-tubes critical values, and a multiplier bootstrap that actually works in simulations. The two-step bias correction for identifiability is sensible, the simulations are extensive, and an R package ships. Second, there is a real, load-bearing gap in the approximation theory. The mapped sieve basis functions in Examples G.1–G.3 all carry a factor sqrt(u'(x)), which decays to zero as |x| -> infinity for the paper's own mappings. So every finite linear combination of these basis functions vanishes at infinity, and so does every sieve approximant. But Assumption 3.1 only requires the mapped function to be smooth on the unit square; it does not require the untransformed m_j(t,x) to decay. A function like m_j(t,x) = sin(2πt) satisfies Assumption 3.1, yet its sup-norm distance to any sieve approximant is at least |sin(2πt)|, because the approximant is zero at infinity. Proposition 3.1, which gives the O(c^{-m1j} + d^{-m2j}) bias rate, is therefore false as stated. Since that bias term is what Theorem 3.2's uniform consistency and Assumption 4.1's under-smoothing condition both rely on, the coverage claims in Theorems 4.1–4.3 are not justified under the stated assumptions. The fix is straightforward: either strengthen Assumption 3.1 to require generalized Schwartz-space decay, or restrict all uniform statements to compact sets and adjust the claims accordingly. Note that the authors do mention decay in Remark 3.1 as a sufficient condition, but it is not part of the assumption, so the paper is internally inconsistent as written. Two smaller soft spots: the condition in Theorem 4.1 that |r| is uniformly bounded below is asserted but not verified for the recommended bases, and several lemmas in the supplement are proved only 'by a discussion similar to,' which is a bit casual given how delicate the uniform Gaussian approximation is. My verdict: this deserves peer review — the gap is identifiable and fixable, and the framework is valuable — but the referee should insist on the corrected assumption or a compact-domain version before the results can be trusted as stated. I'd bring it to a reading group, not to cite it in its current form.

Referee Report

3 major / 5 minor

Summary. The paper proposes sieve estimators for the time-varying nonlinear regression model Y_i = m_0(t_i) + Σ_j m_j(t_i, X_{j,i}) + ε_i under a locally stationary physical-representation framework. It constructs pilot sieve estimators using mapped orthogonal bases on the unbounded covariate domain, corrects them for the identifiability condition E(m_j) = 0, and proves uniform consistency, simultaneous confidence regions over (t,x), a volume-of-tubes expansion for critical values, and a multiplier bootstrap procedure. Numerical simulations and two data applications are included, together with an R package.

Significance. If the main results were correct, the paper would be a substantial contribution: it targets simultaneous inference for a general time-varying nonlinear regression model, provides two Gaussian approximation results for high-dimensional locally stationary affine forms, and offers a practical multiplier bootstrap implementation with an R package. The proof architecture is ambitious and the empirical comparisons against kernel estimators are informative. However, the central sieve approximation result is false under the stated Assumption 3.1, and this defect propagates into the consistency, under-smoothing, and coverage theorems. The contribution is therefore conditional on a substantial correction of the approximation theory and of the assumptions under which it is stated.

major comments (3)

[§3.1.1, Proposition 3.1, Lemma G.2 (Supplement)] The stated approximation result is false under Assumption 3.1. The mapped orthonormal bases in (3.8) and Examples G.1–G.3 have the form φ_j(x) = √(u'(x)) J_j(u(x)); since u'(x) → 0 as |x| → ∞, every φ_j(x) tends to 0 at infinity. Any finite linear combination therefore belongs to C_0(R), the space of functions vanishing at infinity. Assumption 3.1 only requires smoothness of em_j(t, y) = m_j(t, g(2y−1)) on [0,1]^2, which is satisfied, for example, by m_1(t,x) = sin(2πt). For this function, sup_{t,x} |m_1(t,x) − m_{1,c,d}(t,x)| ≥ sup_t |sin(2πt)|, because the approximant tends to 0 as x → ∞. Thus the claimed O(c^{-m_1} + d^{-m_2}) bound in Proposition 3.1 cannot hold under Assumption 3.1. The proof of Lemma G.2 in the supplement equates sup_x |m(t,x) − Σ b_j φ_j(x)| with sup_y |em(t,y) − Σ b_j J_j(y)|; this is not an identity, since the mapped expansion is √(u'(x)) Σ b_j J_j(u(x)), not Σ b_j J_j(u(x)). The proof in Section L.5 merely cites standard compact-domain approximation results and does not repair this gap.
[Theorem 3.2, Assumption 4.1, Theorems 4.1–4.3] Because Proposition 3.1 is false under the stated assumptions, the approximation-bias terms c_j^{-m_{1j}} + d_j^{-m_{2j}} in (3.35) and (4.1) are not justified. These terms are load-bearing: they are used to claim uniform consistency in Theorem 3.2 and to impose the under-smoothing condition in Assumption 4.1, and Theorems 4.1–4.3 then rely on this for the coverage of the simultaneous confidence regions (4.7) and (4.28). Remark 3.1 states a generalized Schwartz-space sufficient condition for the method to work, but that condition is not part of Assumption 3.1, so the theorem statements are internally inconsistent as written.
[§1.2, §3.1.1, Remark 3.1] The counterexample is not a peripheral technicality: the paper's advertised features are that the assumptions are 'mild' and that the method allows 'unbounded domain support.' Under the currently stated assumption, a constant-in-x smooth function such as sin(2πt) is admissible, yet it cannot be uniformly approximated by the proposed mapped sieve space. To repair the manuscript, the authors must add an explicit decay or vanishing-at-infinity condition on m_j(t, ·) (for example, membership in the generalized Schwartz space of order m_{2j} as described in Remark 3.1 and Section J.3), prove Proposition 3.1 from first principles under that condition while accounting for the √(u') weight, and then revisit all downstream rates and coverage results. Merely citing [60] or [12] is not sufficient.

minor comments (5)

[§3.1.1, after (3.7)] The sentence 'Note that {eφ_i} is a sequence of orthogonal basis of the functional space defined on R' should specify the space L^2(R) and the measure involved, since the orthogonality of mapped bases is an L^2 property, not a sup-norm property.
[Lemma G.2 (Supplement)] The displayed equality in the proof uses em_j(t,x) on the right-hand side; the second argument should be y, so that the expression reads em_j(t,y) and em_{j,d}(t,y).
[§1.2 and Figure A caption] There are typos: 'SMIle' should be 'SIMle', and 'generayed' should be 'generated'.
[Equation (3.11) and Section H] The lower index g in the second sum of (3.11) is introduced and then set to 1 for simplicity; the relationship between this convention and the definition of W_1 in (H.2), which uses φ_{ℓ_2+1}, should be clarified in the main text.
[Abstract, §1.2, §3.1.1] The claim that the method 'allow[s] for unbounded domain support' should be reworded after the necessary decay assumption is imposed: the support may be unbounded, but the functions must vanish at infinity with prescribed derivative decay.

Circularity Check

1 steps flagged · score 6.0 of 10

Lemma G.2 assumes the mapped-basis expansion it is used to prove, making the uniform-consistency and SCR coverage chain partially circular.

self definitional [Supplement Section G, Lemma G.2; used in the proof of Proposition 3.1 (Section 3.1.1 and Section L.5)]
"LEMMA G.2. Suppose Assumption 3.1 holds. Then for any fixed t ∈ [0, 1], denote mj,d(t, x) = Σ_{j=1}^d bjφj(x), where we assume that mj(t, x) = Σ_{j=1}^∞ bjφj(x), bj ≡ bj(t). Then for the mapped basis functions in Examples G.1–G.3, we have that sup_{x∈R}|mj(t, x) − mj,d(t, x)| = O(d^{−m2})."

The lemma's conclusion is the sup-norm truncation error of the mapped-basis series. Its premise 'where we assume that mj(t, x) = Σ bjφj(x)' already asserts that m_j is representable by that basis, so the claimed rate is the tail of an assumed identity rather than a derived approximation property. Proposition 3.1 then invokes Lemma G.2 (together with Lemma G.1 and [68]) to conclude the sup-norm rate c_j^{-m1j}+d_j^{-m2j} for every m_j satisfying only Assumption 3.1. The proof of Lemma G.2 also equates sup_x |m_j − Σ b_j φ_j(x)| with sup_y |em_j − Σ b_j J_j(y)|, dropping the sqrt(u'(x)) factor present in the mapped bases of Examples G.1–G.3, so the reduction to the compact-domain result [12] is not the same approximation problem.

full rationale

Most of the paper's statistical derivation is not circular: the OLS construction (3.13), the bias-correction step (3.20)–(3.22), the Gaussian approximation Theorems L.2–L.3, the volume-of-tubes critical-value calculation, and the multiplier bootstrap (Algorithm 1) are developed from the model and dependence assumptions rather than fitted to the targets. The self-citations [23] and [24] supply standard physical-dependence and autoregressive-approximation infrastructure; they do not assume the time-varying sieve conclusion. The identifiability correction estimates the mean of the basis functions, which is a standard identifiability device, not a fitted prediction. However, the approximation foundation is partially circular: Lemma G.2 states its result under the explicit assumption that m_j equals its mapped-basis expansion, and Proposition 3.1's proof leans on that lemma to deliver the c^{-m1j}+d^{-m2j} bias rate for all functions satisfying Assumption 3.1. The mapped basis in Examples G.1–G.3 contains the factor sqrt(u'(x)), so the sup-norm approximation of m_j by Σ b_j φ_j is not the same as the compact-domain approximation of em_j by Σ b_j J_j; the lemma's proof identifies the two by dropping that factor. Thus the central uniform-consistency and SCR-coverage claims inherit an assumed representation rather than a derivation from the stated assumptions. This is a genuine circular step in the proof structure, though it concerns one lemma rather than the whole inferential machinery, so the circularity score is moderate rather than maximal.

Assumptions & free parameters 3 free parameters · 7 assumptions · 0 invented entities

The central theory rests on assumptions about the dependence structure and smoothness of the unknown functions. No new particles, forces, or physical constants are introduced. The ad hoc elements are the under-smoothing condition and the basis norm condition in Theorem 4.1.

free parameters (3)

Sieve dimensions c0, cj, dj = chosen by cross-validation (supplement Section I)
The number of basis functions in t and x controls the bias-variance tradeoff; the data-driven choice via forecast MSE is needed for practical coverage.
Bootstrap block length m = chosen by minimum volatility method with h0=3 (supplement Section I)
The block length in the multiplier bootstrap determines how well the local covariance is approximated; setting it via the volatility criterion is heuristic but standard.
Truncation parameters m and h in Theorems 4.1-4.2 = chosen to minimize Theta and Theta* respectively
These appear only in the theoretical Gaussian approximation bounds (4.13) and (4.16); the rates of convergence depend on optimally balancing them, but they are not used in the algorithm.

assumptions (7)

domain assumption Assumption 2.1: locally stationary processes have physical representation X_j,i = G_j(t_i,F_i), epsilon_i = D(t_i,F_i) with stochastic Lipschitz continuity in t.
Defines the class of processes studied; standard in locally stationary time series literature [74,80,87].
domain assumption Assumption 2.2: physical dependence measures decay as k^{-tau}, tau>1.
Guarantees short-range dependence needed for concentration inequalities and Gaussian approximation.
domain assumption Assumption 3.1: mapped functions em_j(t,y) are C^m with uniformly bounded derivatives on [0,1]^2; equivalent to m_j lying in a generalized Schwartz class.
Essential for the mapped sieve approximation rate in Proposition 3.1; limits the method to functions with rapid decay at infinity.
domain assumption Assumption 3.2: mean functions vartheta_{j,ell}(t) are in C^{n_{j,ell}}([0,1]).
Needed for the bias-correction step to converge; assumed to hold for the chosen bases.
domain assumption Assumption 3.3: integrated long-run covariance matrices Pi and Omega have eigenvalues bounded away from 0 and infinity.
Standard identifiability and regularity condition in sieve least squares regression.
ad hoc to paper Assumption 4.1: under-smoothing, with approximation biases of order O(n^{-epsilon}), epsilon>1/2.
Ensures bias is negligible so that inference focuses on the stochastic error; it requires choosing c_j,d_j in the range (4.2).
ad hoc to paper The vector r(t,x)=b(t,x)-f(t) has norm uniformly bounded away from zero over [0,1]×R.
Stated in Theorem 4.1 to justify the scaling by h_j(t,x); the paper claims common bases satisfy it but does not verify it for the wavelet bases recommended in simulations.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Simultaneous Sieve Estimation and Inference for Time-Varying Nonlinear Time Series Regression." pith.science (2026). https://pith.science/paper/BVRUSFMA

@misc{pith2026250623069,
  author       = {Pith},
  title        = {Pith review of: Simultaneous Sieve Estimation and Inference for Time-Varying Nonlinear Time Series Regression},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BVRUSFMA}},
  note         = {Machine review of arXiv:2506.23069}
}

abstract

In this paper, we investigate time-varying nonlinear time series regression for a broad class of locally stationary time series. First, we propose sieve nonparametric estimators for the time-varying regression functions that achieve uniform consistency. Second, we develop a unified simultaneous inferential theory to conduct both structural and exact form tests on these functions. Additionally, we introduce a multiplier bootstrap procedure for practical implementation. Our methodology and theory require only mild assumptions on the regression functions, allow for unbounded domain support, and effectively address the issue of identifiability for practical interpretation. Technically, we establish sieve approximation theory for 2-D functions in unbounded domains, prove two Gaussian approximation results for affine forms of high-dimensional locally stationary time series, and calculate critical values for the maxima of the Gaussian random field arising from locally stationary time series, which may be of independent interest. Numerical simulations and two data analyses support our results, and we have developed an $\mathtt{R}$ package, $\mathtt{SIMle}$, to facilitate implementation.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

88 extracted references · 79 canonical work pages

[60]

and W ANG , L.-L

S HEN , J. and W ANG , L.-L. (2009). Some recent advances on spectral methods for unbounded domains. Commun. Comput. Phys. 5 195–241

work page 2009
[12]

C HEN , X. (2007). Large Sample Sieve Estimation of Semi-nonparametric Models. Chapter 76 in Handbook of Econometrics, V ol. 6B, James J. Heckman and Edward E. Leamer

work page 2007
[1]

and K ATO, K

B ELLONI , A., C HERNOZHUKOV , V., C HETVERIKOV , D. and K ATO, K. (2015). Some new asymptotic theory for least squares series: pointwise and uniform results. J. Econometrics 186 345–366

2015
[2]

B ENTKUS , V. (2003). On the dependence of the Berry–Esseen bound on dimension. Journal of Statistical Planning and Inference 113 385–402

2003
[3]

B ISHOP , C. M. (2013). Pattern Recognition and Machine Learning . Information science and statistics . Springer

2013
[4]

and T AUCHEN , G

B OLLERSLEV , T., O STERRIEDER , D., S IZOVA, N. and T AUCHEN , G. (2013). Risk and return: Long-run relations, fractional cointegration, and return predictability. Journal of Financial Economics 108 409- 424

2013
[5]

B OYD, J. P. (2001). Chebyshev and Fourier spectral methods, Second ed. Dover Publications, Inc., Mineola, NY

2001
[6]

B OYD, J. P. (2009). Large-degree asymptotics and exponential asymptotics for Fourier, Chebyshev and Hermite coefficients and Fourier transforms. J. Engrg. Math. 63 355–399

2009

Show all 88 references

[7]

B RANDT , M. W. and WANG , L. (2010). Measuring the Time-Varying Risk-Return Relation from the Cross- Section of Equity Returns. Manuscript, Duke University

2010
[8]

T., L IU, W

C AI, T. T., L IU, W. and Z HOU , H. H. (2016). Estimating sparse precision matrix: optimal rates of conver- gence and adaptive estimation. Ann. Statist. 44 455–488

2016
[9]

and L I, R

C AI, Z., F AN, J. and L I, R. (2000). Efficient estimation and inferences for varying-coefficient models. Journal of the American Statistical Association 95 888–902

2000
[10]

and S CAILLET , O

C HAIEB , I., L ANGLOIS , H. and S CAILLET , O. (2021). Factors and risk premia in individual international stock returns. Journal of Financial Economics 141 669-692

2021
[11]

and W U, W

C HEN , L., S METANINA , E. and W U, W. B. (2021). Estimation of nonstationary nonparametric regression model with multiplicative structure. The Econometrics Journal

2021
[13]

C HEN , X. (2013). Penalized Sieve Estimation and Inference of Seminonparametric Dynamic Models: A Selective Review. In Advances in Economics and Econometrics: Tenth World Congress. Econometric Society Monographs 485–544. Cambridge University Press

2013
[14]

and C HRISTENSEN , T

C HEN , X. and C HRISTENSEN , T. M. (2015). Optimal uniform convergence rates and asymptotic normality for series estimators under weak dependence and weak conditions. Journal of Econometrics 188 447- 465

2015
[15]

and W U, W

C HEN , X., X U, M. and W U, W. B. (2013). Covariance and precision matrix estimation for high- dimensional time series. Ann. Statist. 41 2994–3021

2013
[16]

and K ATO, K

C HERNOZHUKOV , V., C HETVERIKOV , D. and K ATO, K. (2013). Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors. The Annals of Statistics 41 2786 – 2819

2013
[17]

and K ALMYKOV , Y

C OFFEY, W. and K ALMYKOV , Y. P. (2012). The Langevin equation: with applications to stochastic prob- lems in physics, chemistry and electrical engineering 27. World Scientific. SIMULTANEOUS SIEVE INFERENCE 65

2012
[18]

C ORSI , F. (2009). A Simple Approximate Long-Memory Model of Realized V olatility.Journal of Financial Econometrics 7 174-196

2009
[19]

and W U, W

D AHLHAUS , R., R ICHTER , S. and W U, W. B. (2019). Towards a general theory for nonlinear locally stationary processes. Bernoulli 25 1013 – 1044

2019
[20]

D AUBECHIES , I. (1988). Orthonormal bases of compactly supported wavelets. Commun. Pure Appl. Math. 41 909-996

1988
[21]

D AUBECHIES , I. (1992). Ten Lectures on Wavelets. SIAM series: CBMS-NSF Regional Conference Series in Applied Mathematics

1992
[22]

Simultaneous sieve estimation and inference for time-varying nonlinear time series regression

D ING , X. and Z HOU , Z. Supplement to "Simultaneous sieve estimation and inference for time-varying nonlinear time series regression"
[23]

and Z HOU , Z

D ING , X. and Z HOU , Z. (2020). Estimation and inference for precision matrices of nonstationary time series. The Annals of Statistics 48 2455 – 2477

2020
[24]

and ZHOU , Z

D ING , X. and ZHOU , Z. (2023). AutoRegressive approximations to nonstationary time series with inference and applications. The Annals of Statistics 51 1207 – 1231

2023
[25]

D URRETT , R. (2019). Probability—theory and examples, fifth ed. Cambridge Series in Statistical and Prob- abilistic Mathematics 49. Cambridge University Press, Cambridge

2019
[26]

and J IANG , J

F AN, J. and J IANG , J. (2005). Nonparametric inferences for additive models. Journal of the American Statistical Association 100 890–907

2005
[27]

and Y AO, Q

F AN, J. and Y AO, Q. (2008). Nonlinear time series: nonparametric and parametric methods . Springer Science & Business Media

2008
[28]

and ZHANG , W

F AN, J. and ZHANG , W. (1999). Statistical estimation in varying coefficient models.The Annals of Statistics 27 1491–1518

1999
[29]

and Z HANG , W

F AN, J. and Z HANG , W. (2000). Simultaneous confidence bands and hypothesis testing in varying- coefficient models. Scand. J. Statist. 27 715–731

2000
[30]

F ANG , X. (2016). A multivariate CLT for bounded decomposable random vectors with the best known rate. J. Theoret. Probab.29 1510–1523

2016
[31]

and R ÖLLIN , A

F ANG , X. and R ÖLLIN , A. (2015). Rates of convergence for multivariate normal approximation with appli- cations to dense graphs and doubly indexed permutation statistics. Bernoulli 21 2157–2189

2015
[32]

F UNARO , D. (1992). Polynomial approximation of differential equations . Lecture Notes in Physics. New Series m: Monographs 8. Springer-Verlag, Berlin

1992
[33]

and M ARCELLINO , M

G HYSELS , E., G UÉRIN , P. and M ARCELLINO , M. (2014). Regime switches in the risk–return trade-off. Journal of Empirical Finance 28 118-138

2014
[34]

G IESSING , A. (2023). Anti-concentration of suprema of Gaussian processes and Gaussian order statistics. arXiv preprint arXiv:2310.12119

2023 arXiv
[35]

G OLUB , G. H. and VAN LOAN, C. F. (2013). Matrix computations, fourth ed. Johns Hopkins Studies in the Mathematical Sciences. Johns Hopkins University Press, Baltimore, MD

2013
[36]

and X IU, D

G U, S., K ELLY, B. and X IU, D. (2020). Empirical Asset Pricing via Machine Learning. The Review of Financial Studies 33 2223-2273

2020
[37]

H ANSEN , B. E. (2008). Uniform convergence rates for kernel estimation with dependent data.Econometric Theory 24 726–748

2008
[38]

H ANSEN , B. E. (2014). Nonparametric sieve regression: least squares, averaging least squares, and cross- validation. In The Oxford handbook of applied nonparametric and semiparametric econometrics and statistics 215–248. Oxford Univ. Press, Oxford

2014
[39]

H OROWITZ , J. (2009). Semiparametric and Nonparametric Methods in Econometrics. Springer

2009
[40]

H OROWITZ , J. L. (2014). Nonparametric Additive Models. In The Oxford Handbook of Applied Nonpara- metric and Semiparametric Econometrics and Statistics Oxford University Press

2014
[41]

and Y OU, J

H U, L., H UANG , T. and Y OU, J. (2019). Estimation and Identification of a Varying-Coefficient Additive Model for Locally Stationary Processes. Journal of the American Statistical Association 114 1191- 1204

2019
[42]

and Y OUNG , N

H UANG , H., M ARCANTOGNINI , S. and Y OUNG , N. (2006). Chain rules for higher derivatives. The Math- ematical Intelligencer 28 61–69

2006
[43]

Z., W U, C

H UANG , J. Z., W U, C. O. and Z HOU , L. (2002). Varying-Coefficient Models and Basis Function Approx- imations for the Analysis of Repeated Measurements. Biometrika 89 111–128

2002
[44]

and J OHNSTONE , I

J OHANSEN , S. and J OHNSTONE , I. M. (1990). Hotelling’s theorem on the volume of tubes: some illustra- tions in simultaneous inference and data analysis. Ann. Statist. 18 652–684

1990
[45]

and W U, W

K ARMAKAR , S., R ICHTER , S. and W U, W. B. (2022). Simultaneous inference for time-varying models. Journal of Econometrics 227 408-428

2022
[46]

and S IEGMUND , D

K NOWLES , M. and S IEGMUND , D. (1989). On Hotelling’s Approach to Testing for a Nonlinear Parameter in Regression. International Statistical Review / Revue Internationale de Statistique 57 205–220. 66

1989
[47]

K RISTENSEN , D. (2009). Uniform convergence rates of kernel estimators with heterogeneous dependent data. Econometric Theory 25 1433–1445

2009
[48]

L ANG , S. (1993). Real and Functional Analysis. Graduate Texts in Mathematics. Springer New York

1993
[49]

and R ACINE , J

L I, Q. and R ACINE , J. S. (2006). Nonparametric Econometrics: Theory and Practice . Economics Books. Princeton University Press

2006
[50]

and W ANG , Q

L INTON , O. and W ANG , Q. (2016). Nonparametric transformation regression with nonstationary data. Econometric Theory 32 1–29

2016
[51]

and L IN, Z

L IU, W. and L IN, Z. (2009). Strong approximation for a class of stationary processes. Stochastic Process. Appl. 119 249–280

2009
[52]

and W U, W

L IU, W., XIAO, H. and W U, W. B. (2013). Probability and moment inequalities under dependence. Statist. Sinica 23 1257–1272

2013
[53]

M ERTON , R. C. (1973). An intertemporal capital asset pricing model. Econometrica 41 867–887

1973
[54]

M EYER , Y. (1990). Ondelettes et opérateurs. I. Actualités Mathématiques. Hermann, Paris Ondelettes

1990
[55]

P., V ON SACHS , R

N ASON , G. P., V ON SACHS , R. and K ROISANDT , G. (2000). Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 62 271–292

2000
[56]

U., M AMMEN , E., L EE, Y

P ARK , B. U., M AMMEN , E., L EE, Y. K. and L EE, E. R. (2015). Varying coefficient regression models: a review and new developments. International Statistical Review 83 36–64

2015
[57]

and Y OU, J

P EI, Y., H UANG , T. and Y OU, J. (2018). Nonparametric fixed effects model for panel data with locally stationary regressors. Journal of Econometrics 202 286-305

2018
[58]

N., W OLF, D

P OLITIS , D. N., W OLF, D. N. P. J. P. R. M., R OMANO , J. P., W OLF, M., B ICKEL , P. J., D IGGLE , P. and FIENBERG , S. (1999). Subsampling. Springer Series in Statistics. Springer New York

1999
[59]

and D AHLHAUS , R

R ICHTER , S. and D AHLHAUS , R. (2019). Cross validation for locally stationary processes. The Annals of Statistics 47 2145 – 2173

2019
[61]

S TONE , C. J. (1982). Optimal global rates of convergence for nonparametric regression. Ann. Statist. 10 1040–1053

1982
[62]

S UN, J. (1993). Tail probabilities of the maxima of Gaussian random fields. Ann. Probab. 21 34–71

1993
[63]

and L OADER , C

S UN, J. and L OADER , C. R. (1994). Simultaneous confidence bands for linear regression and smoothing. Ann. Statist. 22 1328–1345

1994
[64]

and Z HANG , X

S UN, Y., H ONG , Y., L EE, T.-H., W ANG , S. and Z HANG , X. (2021). Time-varying model averaging. J. Econometrics 222 974–992

2021
[65]

S ZEGÖ , G. (1939). Orthogonal Polynomials. American Mathematical Society Colloquium Publications, Vol. 23. American Mathematical Society, New York

1939
[66]

T ALAGRAND , M. (2005). The generic chaining: upper and lower bounds of stochastic processes. Springer Science & Business Media

2005
[67]

T ASAKI , H. (2009). Convergence rates of approximate sums of Riemann integrals. Journal of Approxima- tion Theory 161 477-490

2009
[68]

T IMAN , A. F. (1994). Theory of approximation of functions of a real variable . Dover Publications, Inc., New York Translated from the Russian by J. Berry, Translation edited and with a preface by J. Cossar, Reprint of the 1963 English translation

1994
[69]

VAN HANDEL , R. (2016). Probability in High Dimensions. Lecture notes for APC 550, Princeton Univer- sity

2016
[70]

V OGT, M. (2012). Nonparametric regression for locally stationary time series. The Annals of Statistics 40 2601 – 2633

2012
[71]

V OGT, M. (2015). Testing for structural change in time-varying nonparametric regression models. Econo- metric Theory 31 811–859

2015
[72]

W AINWRIGHT , M. J. (2019). High-dimensional statistics: A non-asymptotic viewpoint . Cambridge Series in Statistical and Probabilistic Mathematics 48. Cambridge University Press, Cambridge

2019
[73]

and Y ANG , L

W ANG , L. and Y ANG , L. (2007). Spline-backfitted kernel smoothing of nonlinear additive autoregression model. The Annals of Statistics 35 2474 – 2503

2007
[74]

W U, W. B. (2005). Nonlinear system theory: another look at dependence. Proc. Natl. Acad. Sci. USA 102 14150–14154

2005
[75]

W U, W. B. (2007). Strong invariance principles for dependent random variables. The Annals of Probability 35 2294 – 2320

2007
[76]

W U, W. B. and Z HOU , Z. (2011). Gaussian approximations for non-stationary multiple time series. Statist. Sinica 21 1397–1413

2011
[77]

Y U, K., P ARK , B. U. and M AMMEN , E. (2008). Smooth backfitting in generalized additive models. The Annals of Statistics 36 228 – 260. SIMULTANEOUS SIEVE INFERENCE 67

2008
[78]

Y UAN, M. (2010). High dimensional inverse covariance matrix estimation via linear programming.J. Mach. Learn. Res. 11 2261–2286

2010
[79]

and W U, W

Z HANG , T. and W U, W. B. (2012). Inference of time-varying regression models. The Annals of Statistics 40 1376 – 1402

2012
[80]

and W U, W

Z HANG , T. and W U, W. B. (2015). Time-varying nonlinear regression models: Nonparametric estimation and model selection. The Annals of Statistics 43 741 – 768

2015
[81]

and W ANG , J.-L

Z HANG , X. and W ANG , J.-L. (2015). Varying-coefficient additive models for functional data. Biometrika 102 15–32

2015
[82]

and W U, W

Z HAO, Z. and W U, W. B. (2008). Confidence bands in nonparametric time series regression. The Annals of Statistics 36 1854 – 1878

2008
[83]

Z HOU , Z. (2013). Heteroscedasticity and autocorrelation robust structural change detection.J. Amer. Statist. Assoc. 108 726–740

2013
[84]

Z HOU , Z. (2014). Nonparametric specification for non-stationary time series regression. Bernoulli 20 78– 108

2014
[85]

Z HOU , Z. (2014). Inference of weighted V -statistics for nonstationary time series and its applications.Ann. Statist. 42 87–114

2014
[86]

Z HOU , Z. (2015). Inference for non-stationary time series regression with or without inequality constraints. J. R. Stat. Soc. Ser. B. Stat. Methodol.77 349–371

2015
[87]

and WU, W

Z HOU , Z. and WU, W. B. (2009). Local linear quantile estimation for nonstationary time series.The Annals of Statistics 37 2696 – 2729

2009
[88]

and W U, W

Z HOU , Z. and W U, W. B. (2010). Simultaneous inference of linear models with time varying coefficients. J. R. Stat. Soc. Ser. B Stat. Methodol.72 513–531

2010

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [60]

and W ANG , L.-L

S HEN , J. and W ANG , L.-L. (2009). Some recent advances on spectral methods for unbounded domains. Commun. Comput. Phys. 5 195–241

work page 2009

[2] [12]

C HEN , X. (2007). Large Sample Sieve Estimation of Semi-nonparametric Models. Chapter 76 in Handbook of Econometrics, V ol. 6B, James J. Heckman and Edward E. Leamer

work page 2007

[3] [1]

and K ATO, K

B ELLONI , A., C HERNOZHUKOV , V., C HETVERIKOV , D. and K ATO, K. (2015). Some new asymptotic theory for least squares series: pointwise and uniform results. J. Econometrics 186 345–366

2015

[4] [2]

B ENTKUS , V. (2003). On the dependence of the Berry–Esseen bound on dimension. Journal of Statistical Planning and Inference 113 385–402

2003

[5] [3]

B ISHOP , C. M. (2013). Pattern Recognition and Machine Learning . Information science and statistics . Springer

2013

[6] [4]

and T AUCHEN , G

B OLLERSLEV , T., O STERRIEDER , D., S IZOVA, N. and T AUCHEN , G. (2013). Risk and return: Long-run relations, fractional cointegration, and return predictability. Journal of Financial Economics 108 409- 424

2013

[7] [5]

B OYD, J. P. (2001). Chebyshev and Fourier spectral methods, Second ed. Dover Publications, Inc., Mineola, NY

2001

[8] [6]

B OYD, J. P. (2009). Large-degree asymptotics and exponential asymptotics for Fourier, Chebyshev and Hermite coefficients and Fourier transforms. J. Engrg. Math. 63 355–399

2009