pith. sign in

arxiv: 2605.28817 · v1 · pith:LWYWEIS6new · submitted 2026-05-27 · 🌌 astro-ph.CO

Fewer simulations, sharper covariances: Reducing mock covariance noise with Zeldovich approximation control variates

Boryana Hadzhiyska , Martin White This is my paper

Pith reviewed 2026-06-29 10:13 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords large-scale structurecovariance estimationmock simulationsZeldovich approximationcontrol variatespower spectrumredshift spaceDESI
0
0 comments X

The pith

Pairing each mock simulation with a cheap Zeldovich approximation realization sharing the same initial conditions reduces the variance of the power spectrum covariance matrix estimate by roughly an order of magnitude on large scales.

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

The paper develops a control-variate method that pairs each expensive mock simulation of large-scale structure with an inexpensive Zeldovich approximation field sharing identical initial conditions. This pairing lets the known statistics of the Zeldovich field subtract out most of the shared sample variance from the covariance estimator. Analytic expressions for the optimal weighting coefficient and the resulting correlation are derived under a Gaussian disconnected approximation. Tests on masked redshift-space lognormal mocks that resemble DESI luminous red galaxies show variance reductions of about a factor of ten below k of 0.05 h per Mpc and factors of two to three at higher wavenumbers. The approach therefore lowers the number of simulations needed to reach a target covariance precision for surveys.

Core claim

By pairing each target mock simulation with a Zeldovich-approximation realization that shares the identical initial conditions, the control-variate estimator removes correlated sample variance from the covariance matrix estimate. Under a Gaussian disconnected approximation the optimal coefficient beta and the resulting correlation rho are given by closed-form expressions involving only the auto- and cross-power spectra. For the monopole the correlation simplifies to the product of the squared cross-correlation coefficients at the two wavenumbers. On DESI-like mocks the method reduces covariance variance by roughly an order of magnitude for wavenumbers below 0.05 h per Mpc and by a factor of

What carries the argument

Zeldovich control variate: a cheap, initial-condition-matched realization whose known statistical properties are used to subtract correlated noise from the target covariance estimator.

If this is right

  • The largest gains occur precisely on the large scales where covariance estimation is most difficult.
  • The computational cost of obtaining accurate covariances drops by a factor of several for typical survey analyses.
  • The analytic expressions allow the expected improvement to be calculated in advance from the cross-power spectrum alone.
  • The method applies immediately to power-spectrum analyses in imaging and spectroscopic surveys.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same pairing idea could be tested on other two-point or higher-order statistics where sample variance is also limiting.
  • Once the cross-correlation coefficient is measured, the expected variance reduction can be forecasted without additional simulations.
  • Extending the approach to multiple control variates of increasing fidelity might yield further gains at modest extra cost.

Load-bearing premise

The target mocks and their Zeldovich counterparts must share a sufficiently large fraction of their sample variance for the subtraction to produce a substantial reduction in estimator noise.

What would settle it

If the measured variance of the covariance-matrix elements obtained from an ensemble of paired mocks equals the variance obtained from the same number of unpaired mocks, the claimed reduction would be falsified.

Figures

Figures reproduced from arXiv: 2605.28817 by Boryana Hadzhiyska, Martin White.

Figure 1
Figure 1. Figure 1: B. Control variate formalism 1. Setup and notation Let Pˆ (s) ℓ (k) denote the power spectrum multipole mea￾sured from the s-th lognormal realization, and let Qˆ (s) ℓ (k) denote the corresponding measurement from the paired Zeldovich realization (same initial conditions). We orga￾nize these into data vectors d (s) and c (s) by concatenat￾ing all (k, ℓ) bins: d (s) i = Pˆ (s) ℓi (ki), c (s) i = Qˆ (s) ℓi (… view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Comparison of the CV weight matrix [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Diagonal comparison of the CV weight [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Comparison of the CV correlation coefficient [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Effect of control variates on redshift-space covariance estimates for the realistic (masked) survey. Left: cor [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Diagonal comparison between the original covariance estimate [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Effective simulation gain (variance reduction) in redshift space with survey mask. Left: analytic prediction using the [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Variance ratio Var[ [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. Accuracy comparison between the original covari [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. Convergence of covariance and precision [PITH_FULL_IMAGE:figures/full_fig_p015_10.png] view at source ↗
read the original abstract

We present a control-variate method for reducing the variance of power spectrum covariance matrix estimates from simulations of large-scale structure. The key idea is to pair each mock simulation with a cheap Zeldovich-approximation realization sharing the same initial conditions, and to use the known statistical properties of the Zeldovich field to remove correlated sample variance from the covariance estimator. Under a Gaussian disconnected approximation, we derive fully analytic expressions for both the optimal control-variate coefficient, $\beta(k,\ell;k',\ell')$, and the corresponding correlation, $\rho(k,\ell;k',\ell')$, in terms of the auto- and cross-power spectra of the target and control fields. In the monopole case, the correlation takes the particularly simple form $\rho(k,k') = r^2(k),r^2(k')$, where $r(k)$ is the standard cross-correlation coefficient between the target and Zeldovich fields, implying that covariance estimation remains highly efficient whenever the two fields are strongly correlated. For masked redshift-space lognormal mocks, resembling Luminous Red Galaxies from the Dark Energy Spectroscopic Instrument (DESI), we find that the control-variate estimator reduces the variance of the covariance matrix by approximately an order of magnitude on large scales, $k \lesssim 0.05\,h\,{\rm Mpc}^{-1}$, precisely where accurate covariance estimation is most challenging. The gains are smaller for higher $k$ but typically accelerate convergence by a factor of 2-3, substantially lowering the computational cost of covariance estimation for current and upcoming large-scale structure surveys. Due to its simplicity, this method is readily implementable in current imaging and spectroscopic surveys (e.g., DESI, Euclid, LSST, PFS, SPHEREx).

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper presents a control-variate method for reducing variance in power-spectrum covariance estimates from large-scale structure mocks. Each full mock is paired with a cheap Zeldovich-approximation realization sharing the same initial conditions; under a Gaussian disconnected approximation, fully analytic expressions are derived for the optimal coefficient β(k,ℓ;k',ℓ') and the correlation ρ(k,ℓ;k',ℓ'). In the monopole this reduces to ρ(k,k')=r²(k)r²(k'), where r(k) is the cross-correlation coefficient between target and Zeldovich fields. Tests on masked redshift-space lognormal mocks resembling DESI LRGs report an order-of-magnitude reduction in covariance-matrix variance for k≲0.05 h Mpc^{-1} and factors of 2–3 at higher k.

Significance. If the gains generalize, the approach would materially lower the number of expensive simulations required for accurate covariance estimation in ongoing and future surveys. The provision of closed-form expressions for β and ρ, together with the especially simple monopole result, constitutes a clear technical strength that distinguishes the work from purely numerical control-variate schemes.

major comments (2)
  1. [Numerical results (lognormal-mock tests)] The central empirical result—an approximately order-of-magnitude reduction in covariance variance for k≲0.05 h Mpc^{-1}—is obtained exclusively from lognormal mocks. Because these fields contain weaker mode coupling and connected four-point contributions than N-body realizations, both the cross-correlation r(k) with the Zeldovich control field and the accuracy of the Gaussian disconnected approximation used to fix β may be systematically higher than would occur in the target application; this directly affects the load-bearing claim that the method substantially lowers computational cost for surveys such as DESI.
  2. [Analytic derivation of β and ρ] The derivation of β(k,ℓ;k',ℓ') and ρ(k,ℓ;k',ℓ') is performed under the Gaussian disconnected approximation. The manuscript does not quantify the residual bias or variance inflation that arises when connected non-Gaussian terms are restored, nor does it test whether the analytic β remains near-optimal once those terms are present; this approximation is load-bearing for the claimed analytic simplicity and efficiency.
minor comments (1)
  1. [Numerical results] A short paragraph describing the precise masking and redshift-space distortion implementation applied to the lognormal fields would improve reproducibility of the reported gains.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive summary and constructive major comments. We respond to each point below.

read point-by-point responses

  1. Referee: [Numerical results (lognormal-mock tests)] The central empirical result—an approximately order-of-magnitude reduction in covariance variance for k≲0.05 h Mpc^{-1}—is obtained exclusively from lognormal mocks. Because these fields contain weaker mode coupling and connected four-point contributions than N-body realizations, both the cross-correlation r(k) with the Zeldovich control field and the accuracy of the Gaussian disconnected approximation used to fix β may be systematically higher than would occur in the target application; this directly affects the load-bearing claim that the method substantially lowers computational cost for surveys such as DESI.

    Authors: We agree that lognormal mocks have weaker mode coupling and connected four-point functions than N-body realizations, so the reported gains may be somewhat optimistic for the most non-Gaussian regimes. However, the largest improvements occur at k ≲ 0.05 h Mpc^{-1}, where the Zeldovich approximation captures the dominant linear and mildly nonlinear contributions that are common to both lognormal and N-body fields; the cross-correlation r(k) is measured directly from the paired realizations rather than assumed. The variance reduction itself is an empirical result from the mocks and does not rely on the Gaussian approximation for its validity. We will add a paragraph in the revised manuscript discussing this limitation of the lognormal tests and outlining how the method can be validated on N-body mocks. revision: yes

  2. Referee: [Analytic derivation of β and ρ] The derivation of β(k,ℓ;k',ℓ') and ρ(k,ℓ;k',ℓ') is performed under the Gaussian disconnected approximation. The manuscript does not quantify the residual bias or variance inflation that arises when connected non-Gaussian terms are restored, nor does it test whether the analytic β remains near-optimal once those terms are present; this approximation is load-bearing for the claimed analytic simplicity and efficiency.

    Authors: The Gaussian disconnected approximation is invoked only to obtain closed-form expressions for the optimal β and the associated ρ; the control-variate estimator itself is implemented using the measured cross-covariance between the target and Zeldovich fields and does not require the approximation. In the lognormal tests (which retain connected four-point contributions), the observed variance reductions closely follow the analytic predictions, indicating that the derived β remains near-optimal even with moderate non-Gaussianity. A precise quantification of residual bias from fully connected terms would require a higher-order perturbative calculation or numerical optimization of β, which lies outside the present scope. We will insert a short clarifying paragraph on the role and limitations of the approximation in the revised manuscript. revision: partial

Circularity Check

0 steps flagged

No circularity; analytic derivation from field statistics is self-contained

full rationale

The paper derives the optimal control-variate coefficient β(k,ℓ;k',ℓ') and correlation ρ(k,ℓ;k',ℓ') under the stated Gaussian disconnected approximation directly from the auto- and cross-power spectra of the target and Zeldovich fields. This follows standard control-variate theory without reducing to a fitted parameter or self-referential definition. The order-of-magnitude variance reduction is an empirical measurement obtained by applying the derived estimator to masked redshift-space lognormal mocks; it is not a prediction that collapses to the input data by construction. No self-citation load-bearing steps, uniqueness theorems, or ansatzes imported from prior work appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of the Gaussian disconnected approximation for the covariance estimation and the assumption that Zeldovich realizations share initial conditions and have computable cross-spectra.

axioms (1)
  • domain assumption Gaussian disconnected approximation
    Invoked to derive analytic expressions for the optimal beta and rho coefficients.

pith-pipeline@v0.9.1-grok · 5857 in / 1349 out tokens · 40139 ms · 2026-06-29T10:13:35.935741+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

63 extracted references · 41 canonical work pages · 27 internal anchors

  1. [1]

    Our simulations, both the target lognormal mocks and the control Zeldovich realizations, share common Gaussian initial conditions

    Initial conditions In this paper, we generate cheap simulations to test our control variates method on lognormal mocks. Our simulations, both the target lognormal mocks and the control Zeldovich realizations, share common Gaussian initial conditions. For each realizations= 1, . . . , N sim, we generate a Gaussian random fieldδ (s) L (k) on a cubic grid of...

    2000

  2. [2]

    Zeldovich approximation The control field is computed from the Zeldovich ap- proximation [39], which provides the leading-order La- 3 grangian perturbation theory prediction for the evolved density field. In the Zeldovich approximation, particles initially at Lagrangian positionqare displaced to Eule- rian position x(q) =q+Ψ(q),(3) where the displacement ...

  3. [3]

    Our procedure follows the implementation ofnbodykit 2 and proceeds as follows

    Lognormal mocks The target simulations are lognormal mock cata- logs [41], which provide a fast way to generate non- Gaussian density fields with approximately correct one- point statistics and two-point clustering. Our procedure follows the implementation ofnbodykit 2 and proceeds as follows. Our chosen redshift,z= 0.5, and bias,b= 2.2 for the lognormal ...

  4. [4]

    Survey mask To test the method in a realistic setting, we apply a survey window function modeled on the DESI South Galactic Cap (SGC) geometry at redshiftz= 0.5. The mask is a binary angular selection function applied to the simulation box; regions outside the observed region in thexycross-section are zeroed, and we additionally cut off 200h −1Mpc from ea...

  5. [5]

    Setup and notation Let ˆP (s) ℓ (k) denote the power spectrum multipole mea- sured from thes-th lognormal realization, and let ˆQ(s) ℓ (k) denote the corresponding measurement from the paired Zeldovich realization (same initial conditions). We orga- nize these into data vectorsd (s) andc (s) by concatenat- ing all (k, ℓ) bins: d(s) i = ˆP (s) ℓi (ki), c (...

  6. [6]

    They arenotmatrices acting on the data vector, but rather scalar coefficients for each element of the covariance ma- trix

    Control variate estimator In our notation,β ij andρ ij are defined element-by- element in the (k, ℓ) space of the covariance matrix. They arenotmatrices acting on the data vector, but rather scalar coefficients for each element of the covariance ma- trix. We could thus denote the elements by a single index, e.g.,α≡ij, but we opt to use the standard double...

  7. [7]

    Xij = ˆΣaa ij , C ij = ˆΣbb ij ,(19) where notationais used to refer to the lognormal mocks, bto the Zeldovich approximation mocks

    Setup The target and control quantities entering the control- variate estimator are elements of sample covariance ma- trices estimated fromN sim paired simulations. Xij = ˆΣaa ij , C ij = ˆΣbb ij ,(19) where notationais used to refer to the lognormal mocks, bto the Zeldovich approximation mocks. The covariance is estimated via ˆΣab ij = 1 Nsim −1 NsimX s=...

  8. [8]

    Covariance of the sample covariance matrix Under the Gaussian approximation, the fields are fully characterised by their two-point functions, and the co- variance of two-point estimators follows from the Isserlis– Wick theorem. For Gaussian fields, the binned power spectrum estimator satisfies Cov ˆP a(ki), ˆP b(kj) ≃ 2 [Pab(ki)]2 Nk,i δij ,(22) whereN k,...

  9. [9]

    Analytic expressions forβandρ Taking the ratio of Eqs. (27) and (28), the (N sim −1) andN k factors cancel exactly, giving the main result: βij = Cov[Xij, C ij] Var[Cij] = [P ab(ki)]2 [P ab(kj)]2 [P bb(ki)]2 [P bb(kj)]2 .(30) Defining α(k)≡ [P ab(k)]2 [P bb(k)]2 ,(31) we write this as βij ≡α(k i)α(k j) (32) The correlation coefficient betweenX ij andC ij ...

  10. [10]

    projection ef- fects

    Extension to multipoles The disconnected derivation above for the monopole relies only on the Gaussian covariance of the underlying bandpower estimators and on the fact that, in theµ- binning used there, differentµ-bins can be treated as ap- proximately independent. For redshift-space multipoles one can proceed analogously, but the crucial difference is t...

    2000

  11. [11]

    The Zeldovich approximation provides an ex- tremely effective control variate: it is cheap to gen- erate and highly accurate on the same large scales where covariance estimation is most numerically challenging

  12. [12]

    We derive an analytic expression for the opti- mal control-variate coefficientβ(k, ℓ;k ′, ℓ′), which is smooth and inexpensive; brute-force validation shows excellent agreement with the analytic pre- diction

  13. [13]

    We validate this dependence against brute-force computations (with a small systematic underpre- diction visible in one figure)

    We derive an analytic expression forρ(k, ℓ;k ′, ℓ′), which depends only on the cross-correlation be- tween the target field and the Zeldovich control field; in the monopole caseρ(k, k ′) =r 2(k)r2(k′). We validate this dependence against brute-force computations (with a small systematic underpre- diction visible in one figure)

  14. [14]

    Despite the covariance involving products of fields, the performance penalty relative to power- spectrum control variates is modest: becauseρfac- torizes asr 2(ki)r2(kj), the resulting covariance per- formance is only about a factor of two worse than the naiver 2 expectation

  15. [15]

    For DESI-like masked redshift-space lognormal mocks, the CV-reduced covariance achieves an O(10) variance-reduction gain on the largest scales, corresponding to an effective∼10 increase in the number of simulations needed for comparable co- variance accuracy

  16. [16]

    Looking ahead, several extensions are natural

    Across a suite of covariance and precision-matrix- sensitive metrics, the CV-reduced estimate is con- sistently closer to the reference “truth” (by about a factor of∼3) and converges faster; whenN sim is close toN data, CV behaves like a shrinkage estima- tor and prevents the onset of ill-conditioning. Looking ahead, several extensions are natural. First,...

  17. [17]

    The Effect of Covariance Estimator Error on Cosmological Parameter Constraints

    S. Dodelson and M. D. Schneider, Phys. Rev. D88, 063537 (2013), 1304.2593

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  18. [18]

    Why your model parameter confidences might be too optimistic -- unbiased estimation of the inverse covariance matrix

    J. Hartlap, P. Simon, and P. Schneider, A&A464, 399 18 (2007), astro-ph/0608064

    work page internal anchor Pith review Pith/arXiv arXiv 2007

  19. [19]

    Putting the Precision in Precision Cosmology: How accurate should your data covariance matrix be?

    A. Taylor, B. Joachimi, and T. Kitching, MNRAS432, 1928 (2013), 1212.4359

    work page internal anchor Pith review Pith/arXiv arXiv 1928

  20. [20]

    W. J. Percival, A. J. Ross, A. G. S´ anchez, L. Samushia, A. Burden, R. Crittenden, A. J. Cuesta, M. V. Magana, M. Manera, F. Beutler, et al., MNRAS439, 2531 (2014), 1312.4841

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  21. [21]

    Parameter inference with estimated covariance matrices

    E. Sellentin and A. F. Heavens, MNRAS456, L132 (2016), 1511.05969

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  22. [22]

    The DESI Experiment Part I: Science,Targeting, and Survey Design

    DESI Collaboration, A. Aghamousa, J. Aguilar, S. Ahlen, S. Alam, L. E. Allen, C. Allende Prieto, J. Annis, S. Bailey, C. Balland, et al., arXiv e-prints arXiv:1611.00036 (2016), 1611.00036

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  23. [23]

    Mellier, Abdurro’uf, J

    Euclid Collaboration, Y. Mellier, Abdurro’uf, J. A. Acevedo Barroso, A. Ach´ ucarro, J. Adamek, R. Adam, G. E. Addison, N. Aghanim, M. Aguena, et al., A&A 697, A1 (2025), 2405.13491

    work page arXiv 2025

  24. [24]

    Cosmology with the SPHEREX All-Sky Spectral Survey

    O. Dor´ e, J. Bock, M. Ashby, P. Capak, A. Cooray, R. de Putter, T. Eifler, N. Flagey, Y. Gong, S. Habib, et al., arXiv e-prints arXiv:1412.4872 (2014), 1412.4872

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  25. [25]

    Extragalactic Science, Cosmology and Galactic Archaeology with the Subaru Prime Focus Spectrograph (PFS)

    M. Takada, R. S. Ellis, M. Chiba, J. E. Greene, H. Ai- hara, N. Arimoto, K. Bundy, J. Cohen, O. Dor´ e, G. Graves, et al., PASJ66, R1 (2014), 1206.0737

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  26. [26]

    LSST Science Collaboration, P. A. Abell, J. Allison, S. F. Anderson, J. R. Andrew, J. R. P. Angel, L. Armus, D. Ar- nett, S. J. Asztalos, T. S. Axelrod, et al., arXiv e-prints arXiv:0912.0201 (2009), 0912.0201

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  27. [27]

    Power Spectrum Correlations Induced by Non-Linear Clustering

    R. Scoccimarro, M. Zaldarriaga, and L. Hui, ApJ527, 1 (1999), astro-ph/9901099

    work page internal anchor Pith review Pith/arXiv arXiv 1999

  28. [28]

    The Growth of Correlations in the Matter Power Spectrum

    A. Meiksin and M. White, MNRAS308, 1179 (1999), astro-ph/9812129

    work page internal anchor Pith review Pith/arXiv arXiv 1999

  29. [29]

    Response Approach to the Matter Power Spectrum Covariance

    A. Barreira and F. Schmidt, J. Cosmology Astropart. Phys.2017, 051 (2017), 1705.01092

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  30. [30]

    Wadekar and R

    D. Wadekar and R. Scoccimarro, Phys. Rev. D102, 123517 (2020), 1910.02914

    work page arXiv 2020

  31. [31]

    Taruya, T

    A. Taruya, T. Nishimichi, and D. Jeong, Phys. Rev. D 103, 023501 (2021), 2007.05504

    work page arXiv 2021

  32. [32]

    Cooray and R

    A. Cooray and R. Sheth, Phys. Rep.372, 1 (2002), astro- ph/0206508

    work page arXiv 2002

  33. [33]

    Covariance of the galaxy angular power spectrum with the halo model

    F. Lacasa, A&A615, A1 (2018), 1711.07372

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  34. [34]

    Simulations of Baryon Acoustic Oscillations II: Covariance matrix of the matter power spectrum

    R. Takahashi, N. Yoshida, M. Takada, T. Matsubara, N. Sugiyama, I. Kayo, A. J. Nishizawa, T. Nishimichi, S. Saito, and A. Taruya, ApJ700, 479 (2009), 0902.0371

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  35. [35]

    L. Blot, M. Crocce, E. Sefusatti, M. Lippich, A. G. S´ anchez, M. Colavincenzo, P. Monaco, M. A. Alvarez, A. Agrawal, S. Avila, et al., MNRAS485, 2806 (2019), 1806.09497

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  36. [36]

    Comparing approximate methods for mock catalogues and covariance matrices I: correlation function

    M. Lippich, A. G. S´ anchez, M. Colavincenzo, E. Se- fusatti, P. Monaco, L. Blot, M. Crocce, M. A. Alvarez, A. Agrawal, S. Avila, et al., MNRAS482, 1786 (2019), 1806.09477

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  37. [37]

    Statistical Analysis of Galaxy Surveys - I. Robust error estimation for 2-point clustering statistics

    P. Norberg, C. M. Baugh, E. Gazta˜ naga, and D. J. Cro- ton, MNRAS396, 19 (2009), 0810.1885

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  38. [38]

    F. G. Mohammad and W. J. Percival, MNRAS514, 1289 (2022), 2109.07071

    work page arXiv 2022

  39. [39]

    A. C. Pope and I. Szapudi, MNRAS389, 766 (2008), 0711.2509

    work page internal anchor Pith review Pith/arXiv arXiv 2008

  40. [40]

    D. J. Paz and A. G. S´ anchez, MNRAS454, 4326 (2015), 1508.03162

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  41. [41]

    Estimating sparse precision matrices

    N. Padmanabhan, M. White, H. H. Zhou, and R. O’Connell, MNRAS460, 1567 (2016), 1512.01241

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  42. [42]

    Non-linear shrinkage estimation of large-scale structure covariance

    B. Joachimi, MNRAS466, L83 (2017), 1612.00752

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  43. [43]

    Precision matrix expansion - efficient use of numerical simulations in estimating errors on cosmological parameters

    O. Friedrich and T. Eifler, MNRAS473, 4150 (2018), 1703.07786

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  44. [44]

    Wadekar, M

    D. Wadekar, M. M. Ivanov, and R. Scoccimarro, Phys. Rev. D102, 123521 (2020), 2009.00622

    work page arXiv 2020

  45. [45]

    S. M. Ross,Simulation(Academic Press, Inc., USA, 2002), 3rd ed., ISBN 0125980531

    2002

  46. [46]

    Chartier, B

    N. Chartier, B. Wandelt, Y. Akrami, and F. Villaescusa- Navarro, MNRAS503, 1897 (2021), 2009.08970

    work page arXiv 2021

  47. [47]

    Chartier and B

    N. Chartier and B. D. Wandelt, MNRAS509, 2220 (2022), 2106.11718

    work page arXiv 2022

  48. [48]

    Chartier and B

    N. Chartier and B. D. Wandelt, MNRAS515, 1296 (2022), 2204.03070

    work page arXiv 2022

  49. [49]

    Kokron, S.-F

    N. Kokron, S.-F. Chen, M. White, J. DeRose, and M. Maus, J. Cosmology Astropart. Phys.2022, 059 (2022), 2205.15327

    work page arXiv 2022

  50. [50]

    DeRose, S.-F

    J. DeRose, S.-F. Chen, N. Kokron, and M. White, J. Cos- mology Astropart. Phys.2023, 008 (2023), 2210.14239

    work page arXiv 2023

  51. [51]

    Control variates from Eulerian and Lagrangian perturbation theory: Application to the bispectrum

    N. Kokron and S.-F. Chen, arXiv e-prints arXiv:2510.07375 (2025), 2510.07375

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  52. [52]

    Hadzhiyska, M

    B. Hadzhiyska, M. J. White, X. Chen, L. H. Garri- son, J. DeRose, N. Padmanabhan, C. Garcia-Quintero, J. Mena-Fern´ andez, S.-F. Chen, H.-J. Seo, et al., The Open Journal of Astrophysics6, 38 (2023), 2308.12343

    work page arXiv 2023

  53. [53]

    Simulation budgeting for hybrid effective field theories

    A. Bartlett, J. DeRose, and M. White, J. Cosmology As- tropart. Phys.2026, 078 (2026), 2510.13962

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  54. [54]

    D. Blas, J. Lesgourgues, and T. Tram, J. Cosmology As- tropart. Phys.2011, 034 (2011), 1104.2933

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  55. [55]

    Y. B. Zel’dovich, A&A5, 84 (1970)

    1970

  56. [56]

    Optimal Weighting Scheme in Redshift-space Power Spectrum Analysis and a Prospect for Measuring the Cosmic Equation of State

    K. Yamamoto, ApJ595, 577 (2003), astro-ph/0208139

    work page internal anchor Pith review Pith/arXiv arXiv 2003

  57. [57]

    Coles and B

    P. Coles and B. Jones, MNRAS248, 1 (1991)

    1991

  58. [58]

    R. Zhou, B. Dey, J. A. Newman, D. J. Eisenstein, K. Dawson, S. Bailey, A. Berti, J. Guy, T.-W. Lan, H. Zou, et al., AJ165, 58 (2023), 2208.08515

    work page arXiv 2023

  59. [59]

    R. Zhou, S. Ferraro, M. White, J. DeRose, N. Sailer, J. Aguilar, S. Ahlen, S. Bailey, D. Brooks, T. Claybaugh, et al., J. Cosmology Astropart. Phys.2023, 097 (2023), 2309.06443

    work page arXiv 2023

  60. [60]

    R. W. Hockney and J. W. Eastwood,Computer Simula- tion Using Particles(1981)

    1981

  61. [61]

    H. A. Feldman, N. Kaiser, and J. A. Peacock, ApJ426, 23 (1994), astro-ph/9304022

    work page internal anchor Pith review Pith/arXiv arXiv 1994

  62. [62]

    Bhatia,Positive Definite Matrices, Princeton Series in Applied Mathematics (Princeton University Press, 2009), ISBN 9781400827787, URLhttps://books

    R. Bhatia,Positive Definite Matrices, Princeton Series in Applied Mathematics (Princeton University Press, 2009), ISBN 9781400827787, URLhttps://books. google.com/books?id=-KIFglY18nYC

    2009

  63. [63]

    R. A. Horn and C. R. Johnson,Matrix Analysis(Cam- bridge University Press, Cambridge, UK, 2012), 2nd ed., ISBN 9780521839402

    2012