arxiv: 2604.18581 · v1 · submitted 2026-04-20 · 🌌 astro-ph.CO

Recognition: unknown

If at First You Don't Succeed, Trispectrum: I. Estimating the Matter Power Spectrum Covariance with Higher-Order Statistics

Samuel Goldstein , Kendrick M. Smith , Utkarsh Giri , Moritz M\"unchmeyer

Authors on Pith no claims yet

Pith reviewed 2026-05-10 03:19 UTC · model grok-4.3

classification 🌌 astro-ph.CO

keywords matter power spectrumcovariance estimationbispectrumtrispectrumnon-Gaussian covariancesuper-sample covarianceN-body simulationslarge-scale structure

0 comments

The pith

Estimators from the squeezed bispectrum and collapsed trispectrum recover the full non-Gaussian matter power spectrum covariance at percent-level accuracy from just 25 simulations.

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

This paper presents a method to estimate the non-Gaussian covariance of the matter power spectrum by directly measuring the response of small-scale power to long-wavelength perturbations. The approach derives explicit estimators for the complete covariance matrix, including super-sample contributions, expressed in terms of the squeezed bispectrum and collapsed trispectrum. When applied to the Quijote N-body simulations, these estimators produce unbiased covariance matrices for scales k greater than or equal to 0.15 h/Mpc that match the precision of traditional sample covariances computed from thousands of simulations. The technique reduces the computational cost of covariance estimation and suggests a route to inferring covariances directly from observational data.

Core claim

We derive estimators for the complete non-Gaussian matter power spectrum covariance, including the super-sample contribution, in terms of the squeezed bispectrum and collapsed trispectrum of the underlying density field. We apply these estimators to the Quijote simulations, and recover unbiased estimates of the small-scale (k ≳ 0.15 h/Mpc) matter power spectrum covariance at the percent level using only 25 simulations.

What carries the argument

Estimators for the power spectrum covariance matrix constructed from measurements of the squeezed bispectrum and collapsed trispectrum, which quantify the response of small-scale modes to long-wavelength density perturbations.

If this is right

Accurate power spectrum covariance estimation becomes feasible with simulation volumes reduced by two orders of magnitude.
Power spectrum analyses can be performed without external covariance matrices by inferring them from the same dataset used for the power spectrum measurement.
Direct comparison between observed higher-order statistics and simulation outputs provides stringent tests of simulation fidelity.
The framework extends in principle to covariance estimation for other two-point statistics beyond the matter power spectrum.

Where Pith is reading between the lines

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

The same response-function approach could be adapted to estimate covariances for galaxy clustering or weak lensing observables if analogous squeezed higher-order spectra can be measured.
In real survey data, residual observational systematics such as mask effects or redshift-space distortions may require additional calibration terms not tested in the simulation-only validation.
Combining the estimator with fast approximate simulations or emulator techniques could further lower the simulation budget while maintaining accuracy.

Load-bearing premise

The response of small-scale power to long-wavelength perturbations is fully captured by the squeezed bispectrum and collapsed trispectrum without significant contributions from higher-order statistics or simulation artifacts.

What would settle it

Direct comparison of the covariance matrix obtained from the bispectrum-trispectrum estimators against the sample covariance computed from an independent suite of several thousand simulations at the same k-range and precision would confirm or refute the claimed percent-level unbiased recovery.

Figures

Figures reproduced from arXiv: 2604.18581 by Kendrick M. Smith, Moritz M\"unchmeyer, Samuel Goldstein, Utkarsh Giri.

**Figure 2.** Figure 2: FIG. 2. Key summary statistics used in this work, measured from the [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. Example marginalized posterior distributions for the [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4. Connected non-Gaussian (left) and super-sample (right) contributions to the matter power spectrum variance estimates. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5. Correlation matrices of the matter power spectrum covariance, showing Gaussian and connected non-Gaussian [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗

**Figure 6.** Figure 6: FIG. 6. Ratio of the mean matter power spectrum averaged over [PITH_FULL_IMAGE:figures/full_fig_p020_6.png] view at source ↗

**Figure 7.** Figure 7: FIG. 7. Sensitivity of the collapsed trispectrum estimator to the disconnected four-point function estimator. We show various [PITH_FULL_IMAGE:figures/full_fig_p021_7.png] view at source ↗

**Figure 8.** Figure 8: FIG. 8. Validation of the estimator pipeline on Gaussian random fields. The top-left panel shows the collapsed trispectrum as a [PITH_FULL_IMAGE:figures/full_fig_p022_8.png] view at source ↗

**Figure 9.** Figure 9: FIG. 9. Impact of the choice of likelihood on the inferred parameters governing the power spectrum covariance. In our fiducial [PITH_FULL_IMAGE:figures/full_fig_p024_9.png] view at source ↗

**Figure 10.** Figure 10: FIG. 10. Correlation matrix of the power spectrum, squeezed bispectrum, and collapsed trispectrum, grouped by soft-mode [PITH_FULL_IMAGE:figures/full_fig_p025_10.png] view at source ↗

**Figure 11.** Figure 11: FIG. 11. Impact of the covariance estimate for the power spectrum, bispectrum, and trispectrum on constraints of the non [PITH_FULL_IMAGE:figures/full_fig_p026_11.png] view at source ↗

**Figure 12.** Figure 12: FIG. 12. Accuracy of the connected non-Gaussian covariance estimated from the squeezed bispectrum and collapsed trispectrum [PITH_FULL_IMAGE:figures/full_fig_p027_12.png] view at source ↗

read the original abstract

We present a method to estimate non-Gaussian power spectrum covariance matrices by directly measuring the response of the small-scale power spectrum to long-wavelength perturbations via bispectrum and trispectrum estimators. Specifically, we derive estimators for the complete non-Gaussian matter power spectrum covariance, including the super-sample contribution, in terms of the squeezed bispectrum and collapsed trispectrum of the underlying density field. We apply these estimators to the Quijote simulations, and recover unbiased estimates of the small-scale ($k\gtrsim 0.15~h/{\rm Mpc}$) matter power spectrum covariance at the percent level using only 25 simulations - comparable to the precision of the sample covariance estimated using 5,000 simulations. This technique significantly reduces the number of simulations needed to estimate power spectrum covariances and opens the possibility of inferring power spectrum covariances directly from survey data, enabling stringent tests of simulations and, potentially, power spectrum analyses that do not rely on external covariance matrices.

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.

Desk Editor's Note private letter to a colleague

This paper maps the full non-Gaussian power spectrum covariance, including super-sample terms, onto squeezed bispectrum and collapsed trispectrum estimators and shows percent-level accuracy on Quijote with only 25 simulations.

read the letter

The punchline is that the authors derive estimators for the complete matter power spectrum covariance from measurements of the squeezed bispectrum and collapsed trispectrum, then recover the sample covariance at the percent level using just 25 Quijote simulations instead of thousands. The derivation constructs the non-Gaussian contributions, including super-sample covariance, directly from these higher-order statistics without circular fitting, and the numerical test on small scales (k ≳ 0.15 h/Mpc) matches the high-statistics result closely enough to be useful. That is the concrete advance: a lower-cost route to reliable covariances that could eventually support data-driven estimates. The work is grounded in the Quijote suite and the results are unbiased within that setup, which gives the claim real weight. The main soft spot is that the validation stays internal to the same simulation suite, so it does not yet test robustness across different cosmologies, volumes, or resolutions. The assumption that the bispectrum and trispectrum exhaust the relevant long-mode responses holds up in the reported tests, but any residual higher-order contributions would only show up in broader checks that are not yet done. The paper does not overclaim generality beyond the demonstrated regime. This is for people who need accurate power spectrum covariances for large-scale structure analyses but cannot run thousands of simulations. Readers focused on covariance modeling or higher-order statistics will get immediate value from the method and the Quijote numbers. It deserves a serious referee because the derivation is new, the evidence is concrete, and the practical payoff is clear even if extra validation would help.

Referee Report

2 major / 2 minor

Summary. The paper derives estimators for the complete non-Gaussian covariance of the matter power spectrum (including super-sample covariance) expressed directly in terms of the squeezed bispectrum and collapsed trispectrum of the density field. These estimators are applied to the Quijote simulation suite, recovering unbiased percent-level estimates of the small-scale (k ≳ 0.15 h/Mpc) power-spectrum covariance matrix using only 25 simulations, matching the precision of the sample covariance computed from 5000 simulations.

Significance. If the mapping from higher-order statistics to the covariance is exact and the numerical results are robust, the work is significant: it reduces the simulation volume required for covariance estimation by two orders of magnitude and opens a path to inferring covariances directly from survey data, which would enable internal consistency tests of simulations and covariance-free power-spectrum analyses.

major comments (2)

[Derivation section (around the estimator definitions)] The central claim of completeness (that the squeezed bispectrum plus collapsed trispectrum exhaust all connected contributions to the four-point function of the power-spectrum estimator, including SSC) is load-bearing for the unbiasedness result. The derivation section should contain an explicit expansion showing that residual five-point and higher connected moments vanish or are negligible in the relevant limits; without this, the percent-level agreement with the 5000-simulation sample covariance remains an internal consistency check rather than a validation of the mapping.
[Numerical results / Quijote application section] In the Quijote application, the comparison is performed within the same simulation suite and volume. To confirm that finite-volume mode-coupling or resolution effects not captured by the collapsed limit do not introduce a systematic offset, the manuscript should report at least one external test (e.g., against an analytic Gaussian + SSC model on a larger box or a different simulation suite).

minor comments (2)

[Estimator definitions] Clarify the precise definition of the 'collapsed' trispectrum limit used in the estimator (e.g., the exact wave-vector configuration and any window-function corrections).
[Results figures] Figure showing the recovered covariance matrix elements would benefit from an additional panel displaying the ratio to the 5000-simulation reference with error bars derived from the 25 realizations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their positive assessment of the significance of our work and for the constructive comments. We address each major point below, indicating where revisions will be made to strengthen the manuscript.

read point-by-point responses

Referee: [Derivation section (around the estimator definitions)] The central claim of completeness (that the squeezed bispectrum plus collapsed trispectrum exhaust all connected contributions to the four-point function of the power-spectrum estimator, including SSC) is load-bearing for the unbiasedness result. The derivation section should contain an explicit expansion showing that residual five-point and higher connected moments vanish or are negligible in the relevant limits; without this, the percent-level agreement with the 5000-simulation sample covariance remains an internal consistency check rather than a validation of the mapping.

Authors: We agree that an explicit expansion of the connected moments would make the completeness of the mapping more rigorous and elevate the numerical results from a consistency check to a direct validation. In the revised manuscript we will add a dedicated subsection (or appendix) that expands the four-point function of the density field in the squeezed and collapsed limits. This will show that five-point and higher connected contributions are suppressed by additional powers of the long-wavelength power spectrum and by volume factors, rendering them negligible on the scales and in the volumes considered. We will also clarify that the relevant object is the two-point function of the power-spectrum estimator. revision: yes
Referee: [Numerical results / Quijote application section] In the Quijote application, the comparison is performed within the same simulation suite and volume. To confirm that finite-volume mode-coupling or resolution effects not captured by the collapsed limit do not introduce a systematic offset, the manuscript should report at least one external test (e.g., against an analytic Gaussian + SSC model on a larger box or a different simulation suite).

Authors: The sample covariance from 5000 independent realizations constitutes the ground-truth covariance within the Quijote volume and resolution. Because the estimator recovers this matrix to the percent level without bias, any finite-volume mode-coupling or resolution effects not captured by the collapsed limit must either be negligible or already included in the measured higher-order statistics. We therefore view the current comparison as a direct validation rather than merely internal. An external test on a larger box would be a useful future extension but is not required to support the claims of the present work. In the revision we will add a brief discussion of this point and of the limitations of the collapsed approximation. revision: partial

Circularity Check

0 steps flagged

No significant circularity: covariance estimators derived from independent higher-order statistics

full rationale

The paper derives estimators for the non-Gaussian power-spectrum covariance (including SSC) by expressing it in terms of the squeezed bispectrum and collapsed trispectrum of the density field. These are independent correlation functions measured directly from the simulations; the mapping does not redefine the covariance in terms of itself, fit parameters to the target covariance, or rely on load-bearing self-citations whose prior results are unverified. The Quijote validation compares the new estimator against the sample covariance from 5000 realizations, but this is an external numerical check rather than a definitional reduction. No step in the provided derivation chain reduces by construction to the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; the approach builds on standard higher-order statistics in cosmology without introducing new ones.

pith-pipeline@v0.9.0 · 5484 in / 1101 out tokens · 30371 ms · 2026-05-10T03:19:13.793337+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

76 extracted references · 67 canonical work pages · 4 internal anchors

[1]

Gaussian covariance The Gaussian covariance can be obtained from Eq.(19) by assuming δ is a Gaussian random field and using Wick’s theorem to simplify the four-point function,i.e., ⟨ ˆP(k b) ˆP(k ′ b)⟩G = 1 Nkb Nk′ b Z p1...p4 Θkb(p1)Θkb(p2)Θk′ b (p3)Θk′ b (p4) ×(2π) 3δ(3) D (p12)(2π)3δ(3) D (p34) h P(p 1)P(p 2)(2π)3δ(3) D (p12)(2π)3δ(3) D (p34) + 2 perm....
[2]

bin-averaged trispec- trum

Connected non-Gaussian covariance In the absence of a survey window function, the connected non-Gaussian contribution is given by the connected con- tribution to the four-point function,i.e., CovcNG kb,k′ b = 1 Nkb Nk′ b Z p1...p4 Θkb(p1)Θkb(p2)Θk′ b (p3)Θk′ b (p4)(2π)3δ(3) D (p12)(2π)3δ(3) D (p34)(2π)3δD(p1234)T(p 1,p 2,p 3,p 4), = 1 V   V 2 Nkb Nk′ b ...
[3]

We include this derivation in Appendix B, and quote the main results here

Super-sample covariance To derive the super-sample covariance, one needs to intro- duce a survey window function. We include this derivation in Appendix B, and quote the main results here. The super-sample covariance can be written as [23] CovSSC kb,k′ b =σ 2 W ∂Pm(kb) ∂δL ∂Pm(k′ b) ∂δL ,(26) 7 where ∂Pm(kb)/∂δL is the response of the isotropic power spec...

2010
[4]

W. J. Percivalet al., Mon. Not. Roy. Astron. Soc.439, 2531 (2014), arXiv:1312.4841 [astro-ph.CO]

work page arXiv 2014
[5]

Mohammed and U

I. Mohammed and U. Seljak, Mon. Not. Roy. Astron. Soc. 445, 3382 (2014), arXiv:1407.0060 [astro-ph.CO]

work page arXiv 2014
[6]

Bertolini, K

D. Bertolini, K. Schutz, M. P. Solon, J. R. Walsh, and K. M. Zurek, Phys. Rev. D93, 123505 (2016), arXiv:1512.07630 [astro-ph.CO]

work page arXiv 2016
[7]

Wadekar and R

D. Wadekar and R. Scoccimarro, Phys. Rev. D102, 123517 (2020), arXiv:1910.02914 [astro-ph.CO]

work page arXiv 2020
[8]

Nicola, C

A. Nicola, C. Garc´ ıa-Garc´ ıa, D. Alonso, J. Dunkley, P. G. Ferreira, A. Slosar, and D. N. Spergel, JCAP03, 067 (2021), arXiv:2010.09717 [astro-ph.CO]

work page arXiv 2021
[9]

Kehagias and A

A. Kehagias and A. Riotto, Nucl. Phys. B873, 514 (2013), arXiv:1302.0130 [astro-ph.CO]

work page arXiv 2013
[10]

Peloso and M

M. Peloso and M. Pietroni, JCAP05, 031 (2013), arXiv:1302.0223 [astro-ph.CO]

work page arXiv 2013
[11]

Goldstein, A

S. Goldstein, A. Esposito, O. H. E. Philcox, L. Hui, J. C. Hill, R. Scoccimarro, and M. H. Abitbol, Phys. Rev. D 106, 123525 (2022), arXiv:2209.06228 [astro-ph.CO]

work page arXiv 2022
[12]

U. Giri, M. M¨ unchmeyer, and K. M. Smith, Phys. Rev. D112, 123544 (2025), arXiv:2305.03070 [astro-ph.CO]

work page arXiv 2025
[13]

Goldstein, O

S. Goldstein, O. H. E. Philcox, J. C. Hill, A. Espos- ito, and L. Hui, Phys. Rev. D109, 043515 (2024), arXiv:2310.12959 [astro-ph.CO]

work page arXiv 2024
[14]

Goldstein, O

S. Goldstein, O. H. E. Philcox, J. C. Hill, and L. Hui, Phys. Rev. D110, 083516 (2024), arXiv:2407.08731 [astro- ph.CO]

work page arXiv 2024
[15]

Goldstein, O

S. Goldstein, O. H. E. Philcox, E. Fondi, and W. R. Coul- ton, Phys. Rev. D112, 083503 (2025), arXiv:2505.13443 [astro-ph.CO]

work page arXiv 2025
[16]

Squeezed Limit non-Gaussianity Estimation with Cosmic Shear

S.-H. Zang and M. M¨ unchmeyer, (2025), arXiv:2512.07295 [astro-ph.CO]

work page internal anchor Pith review Pith/arXiv arXiv 2025
[17]

Barreira and F

A. Barreira and F. Schmidt, JCAP06, 053 (2017), arXiv:1703.09212 [astro-ph.CO]

work page arXiv 2017
[18]

Barreira and F

A. Barreira and F. Schmidt, JCAP11, 051 (2017), arXiv:1705.01092 [astro-ph.CO]

work page arXiv 2017
[19]

Barreira, E

A. Barreira, E. Krause, and F. Schmidt, JCAP06, 015 (2018), arXiv:1711.07467 [astro-ph.CO]

work page arXiv 2018
[20]

Escoffier, M

S. Escoffier, M. C. Cousinou, A. Tilquin, A. Pisani, A. Aguichine, S. de la Torre, A. Ealet, W. Gillard, and E. Jullo, (2016), arXiv:1606.00233 [astro-ph.CO]

work page arXiv 2016
[21]

Euclid preparation. LXXXIX. Accurate and precise data-driven angular power spectrum covariances

K. Naidooet al.(Euclid), (2025), arXiv:2506.09118 [astro- ph.CO]

work page internal anchor Pith review Pith/arXiv arXiv 2025
[22]

Norberg, C

P. Norberg, C. M. Baugh, E. Gaztanaga, and D. J. Croton, Mon. Not. Roy. Astron. Soc.396, 19 (2009), arXiv:0810.1885 [astro-ph]

work page arXiv 2009
[23]

F. G. Mohammad and W. J. Percival, Mon. Not. Roy. Astron. Soc.514, 1289 (2022), arXiv:2109.07071 [astro- ph.CO]

work page arXiv 2022
[24]

Trusovet al., Mon

S. Trusovet al., Mon. Not. Roy. Astron. Soc.527, 9048 (2024), arXiv:2306.16332 [astro-ph.CO]

work page arXiv 2024
[25]

Lacasa and M

F. Lacasa and M. Kunz, Astron. Astrophys.604, A104 (2017), arXiv:1703.03337 [astro-ph.CO]

work page arXiv 2017
[26]

Power Spectrum Super-Sample Covariance

M. Takada and W. Hu, Phys. Rev. D87, 123504 (2013), arXiv:1302.6994 [astro-ph.CO]

work page Pith review arXiv 2013
[27]

Villaescusa-Navarroet al., Astrophys

F. Villaescusa-Navarroet al., Astrophys. J. Suppl.250, 2 (2020), arXiv:1909.05273 [astro-ph.CO]

work page arXiv 2020
[28]

Creminelli and M

P. Creminelli and M. Zaldarriaga, JCAP10, 006 (2004), arXiv:astro-ph/0407059

work page arXiv 2004
[29]

Cheung, A

C. Cheung, A. L. Fitzpatrick, J. Kaplan, and L. Senatore, JCAP02, 021 (2008), arXiv:0709.0295 [hep-th]

work page arXiv 2008
[30]

Tanaka and Y

T. Tanaka and Y. Urakawa, JCAP05, 014 (2011), arXiv:1103.1251 [astro-ph.CO]

work page arXiv 2011
[31]

Creminelli, J

P. Creminelli, J. Nore˜ na, and M. Simonovi´ c, JCAP07, 052 (2012), arXiv:1203.4595 [hep-th]

work page arXiv 2012
[32]

Conformal Symmetries of Adiabatic Modes in Cosmology

K. Hinterbichler, L. Hui, and J. Khoury, JCAP08, 017 (2012), arXiv:1203.6351 [hep-th]

work page Pith review arXiv 2012
[33]

On Soft Limits of Inflationary Correlation Functions

V. Assassi, D. Baumann, and D. Green, JCAP11, 047 (2012), arXiv:1204.4207 [hep-th]

work page Pith review arXiv 2012
[34]

Kehagias and A

A. Kehagias and A. Riotto, Nucl. Phys. B864, 492 (2012), arXiv:1205.1523 [hep-th]

work page arXiv 2012
[35]

Pajer, F

E. Pajer, F. Schmidt, and M. Zaldarriaga, Phys. Rev. D 88, 083502 (2013), arXiv:1305.0824 [astro-ph.CO]

work page arXiv 2013
[36]

An Infinite Set of Ward Identities for Adiabatic Modes in Cosmology

K. Hinterbichler, L. Hui, and J. Khoury, JCAP01, 039 (2014), arXiv:1304.5527 [hep-th]

work page Pith review arXiv 2014
[37]

W. D. Goldberger, L. Hui, and A. Nicolis, Phys. Rev. D 87, 103520 (2013), arXiv:1303.1193 [hep-th]

work page Pith review arXiv 2013
[38]

Baldauf, M

T. Baldauf, M. Mirbabayi, M. Simonovi´ c, and M. Zaldar- riaga, Phys. Rev. D92, 043514 (2015), arXiv:1504.04366 [astro-ph.CO]

work page arXiv 2015
[39]

Bravo, S

R. Bravo, S. Mooij, G. A. Palma, and B. Pradenas, JCAP 05, 025 (2018), arXiv:1711.05290 [astro-ph.CO]

work page arXiv 2018
[40]

L. Hui, A. Joyce, and S. S. C. Wong, JCAP02, 060 (2019), arXiv:1811.05951 [hep-th]

work page Pith review arXiv 2019
[41]

Chiang and A

C.-T. Chiang and A. Slosar, JCAP07, 049 (2018), arXiv:1804.02753 [astro-ph.CO]

work page arXiv 2018
[42]

Masaki, T

S. Masaki, T. Nishimichi, and M. Takada, Mon. Not. Roy. Astron. Soc.496, 483 (2020), arXiv:2003.10052 [astro- ph.CO]

work page arXiv 2020
[43]

Large-Scale Galaxy Bias

V. Desjacques, D. Jeong, and F. Schmidt, Phys. Rept. 733, 1 (2018), arXiv:1611.09787 [astro-ph.CO]

work page Pith review arXiv 2018
[44]

Lewis, JCAP10, 026 (2011), arXiv:1107.5431 [astro- ph.CO]

A. Lewis, JCAP10, 026 (2011), arXiv:1107.5431 [astro- ph.CO]

work page arXiv 2011
[45]

Kenton and D

Z. Kenton and D. J. Mulryne, JCAP10, 035 (2016), arXiv:1605.03435 [astro-ph.CO]

work page arXiv 2016
[46]

W. R. Coulton, O. H. E. Philcox, and F. Villaescusa- Navarro, (2023), arXiv:2306.11782 [astro-ph.CO]

work page arXiv 2023
[47]

K. M. Smith, L. Senatore, and M. Zaldarriaga, (2015), arXiv:1502.00635 [astro-ph.CO]

work page arXiv 2015
[48]

D. Shen, E. Schaan, and S. Ferraro, (2024), arXiv:2402.04309 [astro-ph.CO]

work page arXiv 2024
[49]

Hanson, A

D. Hanson, A. Challinor, G. Efstathiou, and P. Bielewicz, Phys. Rev. D83, 043005 (2011), arXiv:1008.4403 [astro- 14 ph.CO]

work page arXiv 2011
[50]

Namikawa, D

T. Namikawa, D. Hanson, and R. Takahashi, Mon. Not. Roy. Astron. Soc.431, 609 (2013), arXiv:1209.0091 [astro- ph.CO]

work page arXiv 2013
[51]

Y. Li, W. Hu, and M. Takada, Phys. Rev. D89, 083519 (2014), arXiv:1401.0385 [astro-ph.CO]

work page arXiv 2014
[52]

Pylians: Python libraries for the analysis of numerical simulations,

F. Villaescusa-Navarro, “Pylians: Python libraries for the analysis of numerical simulations,” Astrophysics Source Code Library, record ascl:1811.008 (2018), ascl:1811.008

2018
[53]

W. R. Coulton, F. Villaescusa-Navarro, D. Jamieson, M. Baldi, G. Jung, D. Karagiannis, M. Liguori, L. Verde, and B. D. Wandelt, Astrophys. J.943, 64 (2023), arXiv:2206.01619 [astro-ph.CO]

work page arXiv 2023
[54]

The Cosmic Linear Anisotropy Solving System (CLASS) I: Overview

J. Lesgourgues, (2011), arXiv:1104.2932 [astro-ph.IM]

work page Pith review arXiv 2011
[55]

Sellentin and A

E. Sellentin and A. F. Heavens, Mon. Not. Roy. Astron. Soc.456, L132 (2016), arXiv:1511.05969 [astro-ph.CO]

work page arXiv 2016
[56]

Biagetti, L

M. Biagetti, L. Castiblanco, J. Nore˜ na, and E. Sefusatti, JCAP09, 009 (2022), arXiv:2111.05887 [astro-ph.CO]

work page arXiv 2022
[57]

emcee: The MCMC Hammer

D. Foreman-Mackey, D. W. Hogg, D. Lang, and J. Goodman, Publ. Astron. Soc. Pac.125, 306 (2013), arXiv:1202.3665 [astro-ph.IM]

work page internal anchor Pith review arXiv 2013
[58]

Gelman and D

A. Gelman and D. B. Rubin, Statist. Sci.7, 457 (1992)

1992
[59]

Y. Li, S. Singh, B. Yu, Y. Feng, and U. Seljak, JCAP 01, 016 (2019), arXiv:1811.05714 [astro-ph.CO]

work page arXiv 2019
[60]

M. Maus, A. Baleato Lizancos, M. White, A. de Mattia, and S. Chen, (2026), arXiv:2602.12343 [astro-ph.CO]

work page arXiv 2026
[61]

The Atacama Cosmology Telescope: DR6 Power Spectra, Likelihoods and $\Lambda$CDM Parameters

T. Louiset al.(Atacama Cosmology Telescope), JCAP 11, 062 (2025), arXiv:2503.14452 [astro-ph.CO]

work page internal anchor Pith review arXiv 2025
[62]

Chaussidon, C

E. Chaussidonet al., JCAP06, 029 (2025), arXiv:2411.17623 [astro-ph.CO]

work page arXiv 2025
[63]

O. H. E. Philcox, Phys. Rev. D103, 103504 (2021), arXiv:2012.09389 [astro-ph.CO]

work page arXiv 2021
[64]

O. H. E. Philcox, Phys. Rev. D104, 123529 (2021), arXiv:2107.06287 [astro-ph.CO]

work page arXiv 2021
[65]

O. H. E. Philcox, Phys. Rev. D107, 123516 (2023), arXiv:2303.08828 [astro-ph.CO]

work page arXiv 2023
[66]

O. H. E. Philcox, Phys. Rev. D108, 063506 (2023), arXiv:2306.03915 [astro-ph.CO]

work page arXiv 2023
[67]

O. H. E. Philcox and T. Fl¨ oss, Phys. Rev. D112, 063507 (2025), arXiv:2404.07249 [astro-ph.CO]

work page arXiv 2025
[68]

Gualdi, S

D. Gualdi, S. Novell-Masot, H. Gil-Marin, and L. Verde, JCAP01, 015 (2021), arXiv:2009.02290 [astro-ph.CO]

work page arXiv 2021
[69]

Tegmark, A

M. Tegmark, A. J. S. Hamilton, M. A. Strauss, M. S. Vogeley, and A. S. Szalay, Astrophys. J.499, 555 (1998), arXiv:astro-ph/9708020

work page arXiv 1998
[70]

A. J. S. Hamilton, Lect. Notes Phys.665, 415 (2008), arXiv:astro-ph/0503603

work page arXiv 2008
[71]

A. J. S. Hamilton, Lect. Notes Phys.665, 433 (2008), arXiv:astro-ph/0503604

work page arXiv 2008
[72]

Taylor and B

A. Taylor and B. Joachimi, Mon. Not. Roy. Astron. Soc. 442, 2728 (2014), arXiv:1402.6983 [astro-ph.CO]. 15 Appendix A: A review of 3D trispectrum estimators In this appendix, we provide additional details regarding the three-dimensional (3D) trispectrum estimator used in this work, based on Refs. [ 11, 43] and implemented in the PNGolin package.18 Our goa...

work page arXiv 2014
[73]

Collapsed trispectrum estimator withQ b ̸= 0 An estimator for the total four-point function,i.e., including disconnected contributions, is ˆTtot.(kb1 , kb2 , kb3 , kb4 , Qb)∝ Z P ΘQb(P) Z p1,...,p4 Θkb1 (p1)Θkb2 (p2)Θkb3 (p3)Θkb4 (p4)δ p1 δp2 δp3 δp4 (A1) (2π)3δ(3) D (p12 −P)(2π) 3δ(3) D (p34 +P) , The normalization of the estimator can be computed by re-...
[74]

Exact collapsed limit estimator (Q b = 0) In the exact collapsed limit, the estimator can be derived by setting P = 0 and taking kb ≡k b1 = kb2 and k′ b ≡k b3 = kb4 in Eq. (A1). ˆTtot.(kb, k′ b)∝ Z p1,...,p4 Θkb(p1)Θkb(p2)Θk′ b (p3)Θk′ b (p4)δ p1 δp2 δp3 δp4(2π)3δ(3) D (p12)(2π)3δ(3) D (p34) , = Z d3x(δ kb(x))2 × Z d3y(δ k′ b (y))2 .(A10) Therefore, in th...
[75]

Connected non-Gaussian covariance The connected non-Gaussian covariance, CovcNG kb,k′ b , arises from small-scale mode coupling within the survey volume. Since the trispectrum varies on scales k≫k f ∼V −1/3, it can be pulled outside the window function integrals, giving CovcNG kb,k′ b = 1 Nkb Nk′ b Z p,p′ Θkb(p)Θk′ b (p′)T(p,−p,p ′,−p ′)×   Z q1...q4 4Y...
[76]

To derive this contribution starting with Eq

Super-sample covariance The super-sample covariance refers to the covariance sourced by the mode coupling of modes within the survey volume to modes that are larger than the survey volume. To derive this contribution starting with Eq. (B8), we make the change of variablesu≡p−q 1 andv≡p ′ −q 3 CovNG kb,k′ b = 1 Nkb Nk′ b Z u,v Z q1...q4 Θkb(|u+q 1|)Θk′ b (...