betaFeedback

WorksMathematics

A High-Dimensional LM Test for Cross-Sectional Independence with Serially Correlated Data

Abstract

We develop a high-dimensional Lagrange multiplier (LM) test for cross-sectional independence in Gaussian panels with a common stationary, short-memory temporal covariance matrix. The statistic is based on the trace of the square of R, where R is the sample correlation matrix. When the cross-sectional and time dimensions grow proportionally, we derive explicit expansions for the null mean and variance in terms of temporal spectral moments. Random diagonal normalization produces a non-negligible centering correction, while squared sample correlations sharing a cross-sectional unit contribute an additional variance term. Both corrections vanish under temporal independence. We establish a central limit theorem under the null, consistency against fixed alternatives, and a Gaussian-shift limit with a detection boundary under local alternatives. Monte Carlo experiments examine the size and power of the oracle test and feasible versions that estimate the temporal spectral moments. The asymptotic theory applies to the oracle statistic; feasible procedures are evaluated numerically.

JEL Classification Numbers: C12, C31, C38

Corresponding author and proofs: Chihwa Kao, Department of Economics, University of Connecticut, Storrs, CT 06269, USA; e-mail: [email protected].

Introduction

The LM test of Breusch and Pagan (1980) aggregates squared sample correlations and has a chi-squared null limit when the cross-sectional dimension is fixed and the time dimension grows. In high dimensions, Schott (2005) derives a Gaussian limit for the sum of squared correlations when both dimensions increase proportionally under temporally independent Gaussian sampling. Baltagi, Feng, and Kao (2012) extend high-dimensional LM testing to fixed-effects panel regressions, and Huang, Li, and Yao (2025) develop unified and robust LM-type tests for large panels; both assume serially independent errors.

The existing literature does not establish the centering and scaling of the quadratic LM statistic under serial correlation when the cross-sectional and time dimensions grow proportionally. Baltagi, Kao, and Peng (2016) document size distortions of existing cross-sectional correlation tests under serial dependence and develop a modified CD test; Hounyo, Kao, and Kim (2025) also study CD-type statistics. Those procedures aggregate correlations linearly, whereas the LM statistic aggregates their squares. Chen and Liu (2018) consider a separable Gaussian model but test independence in the temporal dimension. Thus, these results do not directly provide a high-dimensional null distribution for the quadratic LM statistic with serially correlated observations.

We study an n×Tn \times T Gaussian panel whose cross-sectional units are independent under the null and share a stationary, short-memory Toeplitz temporal covariance matrix Γ\Gamma, normalized by Tr⁡(Γ)/T=1\operatorname{Tr}(\Gamma)/T = 1. Write R=(rij)R = (r_{ij}) for the sample correlation matrix formed using the maintained zero population means. Let λ=T/n\lambda= T/n denote the dimension ratio, with λ→λ0∈(0,∞)\lambda\to\lambda_0 \in(0,\infty). We consider the statistic

Tr⁡(R2)=n+2∑i<jrij2.\operatorname{Tr}(R^2) = n + 2\sum_{i<j} r_{ij}^2.

Although this is a linear spectral statistic (LSS), existing CLTs for high-dimensional sample correlation matrices (Gao, Han, Pan, and Yang, 2017; Yin, Zheng, and Zou, 2023; Li, Pan, Xie, and Zhou, 2024) address temporally independent sampling and cannot be applied directly to our setting.

The contribution of this paper is to show how to center and scale Tr⁡(R2)\operatorname{Tr}(R^{2}) under serial correlation. Let τk,T\tau_{k,T} denote the normalized temporal spectral moments of Γ\Gamma, and let KTK_{T} be the function of τ2,T,τ3,T,τ4,T\tau_{2,T},\tau_{3,T},\tau_{4,T} defined in § 2.1. We show that

E{Tr⁡(R2)}=n+n−1λτ2,T+4(n−1)nλ2(τ2,T2−τ3,T)+O(n−1),\mathbb{E}\{\operatorname{Tr}(R^{2})\}=n+\frac{n-1}{\lambda}\tau_{2,T}+\frac{4(n-1)}{n\lambda^{2}}(\tau_{2,T}^{2}-\tau_{3,T})+O(n^{-1}),
Var⁡{Tr⁡(R2)}=4τ2,T2λ2+8KTλ3+O(n−1),\operatorname{Var}\{\operatorname{Tr}(R^{2})\}=\frac{4\tau_{2,T}^{2}}{\lambda^{2}}+\frac{8K_{T}}{\lambda^{3}}+O(n^{-1}),

and that Tr⁡(R2)\operatorname{Tr}(R^{2}), centered and scaled accordingly, is asymptotically standard normal. The centering and scaling depend on the temporal dependence only through τ2,T\tau_{2,T}, τ3,T\tau_{3,T}, and τ4,T\tau_{4,T}, and reduce to the known-mean version of Schott (2005) when Γ=IT\Gamma=I_{T}. Whereas the LSS literature relies on random matrix theory, we obtain these moments directly by Wick’s theorem. This is the key step: once the centering and scaling are known, asymptotic normality is established by a martingale central limit theorem. Details are given in § 3.

The null CLT concerns the oracle statistic, which uses the population spectral moments τk,T\tau_{k,T}; feasible versions based on estimated moments are evaluated in the Monte Carlo experiments, which also illustrate the size distortions that arise when serial correlation is ignored.

We also characterize the behavior of the test under alternatives. We establish consistency against fixed alternatives and a Gaussian shift under local alternatives, with aggregate signal ∥R∗−In∥F2\lVert R_{*}-I_{n}\rVert_{F}^{2}, where R∗R_{*} is the population cross-sectional correlation matrix. This yields an explicit local-power expression and a detection boundary for the statistic under the alternatives considered here; it is not asserted to be a minimax boundary over all tests.

§ 2 presents the model and statistic, § 3 establishes the null expansions and CLT, § 4 studies power, and § 5 reports simulations. § 6 concludes. Proofs of the null results (Lemmas 2 and 3 and Theorem 1) are in the Appendix; the remaining proofs, including those of the power results, are in the Online Supplement.

Model and Test Statistic

Model Specification

Let X=(xit)∈Rn×TX=(x_{it})\in\mathbb{R}^{n\times T} denote observations on nn cross-sectional units over TT periods.

Assume the separable Gaussian model

X=U1/2ZΓ1/2,X=U^{1/2}Z\Gamma^{1/2},

where U∈Rn×nU\in\mathbb{R}^{n\times n} is a positive-definite cross-sectional covariance matrix, Γ∈RT×T\Gamma\in\mathbb{R}^{T\times T} is a temporal covariance matrix, and ZZ has independent N(0,1)N(0,1) entries. We use symmetric matrix square roots. Then vec⁡(X)∼N(0,Γ⊗U)\operatorname{vec}(X)\sim N(0,\Gamma\otimes U) under column-wise vectorization, and Cov⁡(xit,xjs)=UijΓts\operatorname{Cov}(x_{it},x_{js})=U_{ij}\Gamma_{ts}. We write λ=T/n\lambda=T/n and impose proportional growth in Assumption 3.

The temporal covariance matrix Γ=ΓT=(γ∣t−s∣)t,s=1T\Gamma=\Gamma_T=(\gamma_{\lvert t-s\rvert})_{t,s=1}^T, with γ0=1\gamma_0=1, is stationary and short-memory; the precise conditions are given in Assumption 2. We suppress the subscript TT except where the dependence on TT matters. Define the normalized temporal spectral moments by

τk,T=1TTr⁡(Γk)=1T∑t=1Tνt,Tk,k=1,2,…,\tau_{k,T}=\frac{1}{T}\operatorname{Tr}(\Gamma^k)=\frac{1}{T}\sum_{t=1}^{T}\nu_{t,T}^{k},\qquad k=1,2,\ldots,

where ν1,T,…,νT,T\nu_{1,T},\ldots,\nu_{T,T} are the eigenvalues of Γ\Gamma. Since γ0=1\gamma_0=1, τ1,T=1\tau_{1,T}=1. The normalization γ0=1\gamma_0=1 is without loss of generality for the sample correlation matrix, which is invariant to positive scalar rescaling of Γ\Gamma. Define

KT=τ4,T−2τ2,Tτ3,T+τ2,T3,K=lim⁡T→∞KT=τ4−2τ2τ3+τ23,K_T=\tau_{4,T}-2\tau_{2,T}\tau_{3,T}+\tau_{2,T}^{3},\qquad K=\lim_{T\to\infty}K_T=\tau_4-2\tau_2\tau_3+\tau_2^3,

where τk=lim⁡T→∞τk,T\tau_k=\lim_{T\to\infty}\tau_{k,T}.

Hypotheses and Test Statistic

Let xi=(xi1,…,xiT)′∈RTx_i=(x_{i1},\ldots,x_{iT})^\prime\in\mathbb{R}^T be the iith row of XX written as a column vector. We test cross-sectional independence. Under Gaussianity, this is equivalent to UU being diagonal, or to the population correlation matrix R∗=diag⁡(U)−1/2Udiag⁡(U)−1/2R_*=\operatorname{diag}(U)^{-1/2}U\operatorname{diag}(U)^{-1/2} being InI_n. Because the sample correlation matrix is invariant to positive row-wise rescaling, we use the normalization U=InU = I_n when deriving the null distribution. The null hypothesis is therefore H0:R∗=InH_0 : R_* = I_n, with U=InU = I_n as its normalized representation.

With the population mean specified as zero, define the uncentered sample covariance and correlation matrices by

S=1TXX′,R=diag⁡(S)−1/2Sdiag⁡(S)−1/2.S = \frac{1}{T}XX^{\prime}, \qquad R = \operatorname{diag}(S)^{-1/2} S \operatorname{diag}(S)^{-1/2}.

In particular,

rij=∑t=1Txitxjt(∑t=1Txit2)1/2(∑t=1Txjt2)1/2,Tr⁡(R2)=n+2∑i<jrij2.r_{ij} = \frac{\sum_{t=1}^{T} x_{it}x_{jt}}{\left(\sum_{t=1}^{T} x_{it}^{2}\right)^{1/2}\left(\sum_{t=1}^{T} x_{jt}^{2}\right)^{1/2}}, \qquad\operatorname{Tr}(R^{2}) = n + 2\sum_{i<j} r_{ij}^{2}.

This identity expresses the quadratic spectral statistic as a sum of squared off-diagonal sample correlations, the basis of the LM-type test.

Under H0H_0 and Γ=IT\Gamma= I_T, the known-mean version of the standardized statistic of Schott (2005) is

LMS=∑i<jrij2−n(n−1)2Tn(n−1)(T−1)T2(T+2)=Tr⁡(R2)−n−n(n−1)T2n(n−1)(T−1)T2(T+2).LM_S = \frac{\sum_{i<j} r_{ij}^{2} - \frac{n(n-1)}{2T}}{\sqrt{\frac{n(n-1)(T-1)}{T^{2}(T+2)}}} = \frac{\operatorname{Tr}(R^{2}) - n - \frac{n(n-1)}{T}}{2\sqrt{\frac{n(n-1)(T-1)}{T^{2}(T+2)}}}.

Assumptions

The null results below use Assumptions 1–3. The analysis of alternatives additionally invokes Assumption 4.

Assumption 1 The entries ZitZ_{it} are independent and identically distributed N(0,1)\mathrm{N}(0,1) random variables.

References

  1. [1]Davidson, K. R., & Szarek, S. J. (2001). Local operator theory, random matrices and Banach spaces. In W. B. Johnson & J. Lindenstrauss (Eds.), Handbook of the geometry of Banach spaces (Vol. 1, pp. 317–366). North-Holland.DOI
  2. [2]Hounyo, U., Kao, C., & Kim, M. S. (2026a). Limiting spectral distribution for large sample correlation matrices under temporal dependence [Working paper]. University of Connecticut.
  3. [3]Isserlis, L. (1918). On a formula for the product-moment coefficient of any order of a normal frequency distribution in any number of variables. Biometrika, 12(1–2), 134–139. doi:10.1093/biomet/12.1-2.134DOI
  4. [4]Rudelson, M., & Vershynin, R. (2013). Hanson–Wright inequality and sub-Gaussian concentration. Electronic Communications in Probability, 18, Article 82. doi:10.1214/ECP.v18-2865DOI

Paper details

vlnc:W_afyk5gnt
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC0-1.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Claude, Chat-GPT

View statement
Looking for related papers…
Contents