Putting the precision in precision cosmology: How accurate should your data covariance matrix be?

Cosmological parameter estimation requires that the likelihood function of the data is accurately known. Assuming that cosmological large-scale structure power spectra data are multivariate Gaussian-distributed, we show the accuracy of parameter estimation is limited by the accuracy of the inverse data covariance matrix - the precision matrix. If the data covariance and precision matrices are estimated by sampling independent realisations of the data, their statistical properties are described by the Wishart and Inverse-Wishart distributions, respectively. Independent of any details of the survey, we show that the fractional error on a parameter variance, or a Figure-of-Merit, is equal to the fractional variance of the precision matrix. In addition, for the only unbiased estimator of the precision matrix, we find that the fractional accuracy of the parameter error depends only on the difference between the number of independent realisations and the number of data points, and so can easily diverge. For a 5% error on a parameter error and N_D >100 we need N_S > N_D realisations and a fractional accuracy of 10^4 -10^6 this approach will be problematic. We discuss possible ways to relax these conditions: improved theoretical modelling; shrinkage methods; data-compression; simulation and data resampling methods.

[1]  I. K. Wehus,et al.  MARGINAL DISTRIBUTIONS FOR COSMIC VARIANCE LIMITED COSMIC MICROWAVE BACKGROUND POLARIZATION DATA , 2008, 0806.3074.

[2]  Max Tegmark,et al.  Karhunen-Loève Eigenvalue Problems in Cosmology: How Should We Tackle Large Data Sets? , 1996, astro-ph/9603021.

[3]  A. Taylor,et al.  Simulating the effect of non-linear mode coupling in cosmological parameter estimation , 2011, 1103.3245.

[4]  P. Schneider,et al.  COSEBIs: Extracting the full E-/B-mode information from cosmic shear correlation functions , 2010, 1002.2136.

[5]  P. Schneider,et al.  Cosmic shear tomography and efficient data compression using COSEBIs , 2012, 1201.2669.

[6]  L. R. Haff.  ESTIMATION OF THE INVERSE COVARIANCE MATRIX: RANDOM MIXTURES OF THE INVERSE WISHART MATRIX AND THE IDENTITY , 1979 .

[7]  J. Loh.  A Valid and Fast Spatial Bootstrap for Correlation Functions , 2008, 0805.2325.

[8]  A. Heavens,et al.  The real-space power spectrum of IRAS galaxies on large scales and the redshift distortion , 1995 .

[9]  Likelihood techniques for the combined analysis of CMB temperature and polarization power spectra , 2006, astro-ph/0604547.

[10]  M. Zaldarriaga,et al.  Power Spectrum Correlations Induced by Nonlinear Clustering , 1999, astro-ph/9901099.

[11]  A. N. Taylor,et al.  Probing dark energy with the shear-ratio geometric test , 2006, astro-ph/0606416.

[12]  Antonio Padilla,et al.  Modified Gravity and Cosmology , 2011, 1106.2476.

[13]  U. Toronto,et al.  Estimating the power spectrum of the cosmic microwave background , 1997, astro-ph/9708203.

[14]  The growth of correlations in the matter power spectrum , 1998, astro-ph/9812129.

[15]  N. Kaiser.  Clustering in real space and in redshift space , 1987 .

[16]  Anja Vogler,et al.  An Introduction to Multivariate Statistical Analysis , 2004 .

[17]  J. Shao,et al.  A General Theory for Jackknife Variance Estimation , 1989 .

[18]  Shinji Tsujikawa,et al.  Dynamics of dark energy , 2006 .

[19]  S. Bridle,et al.  Probing dark energy with cluster counts and cosmic shear power spectra: including the full covariance , 2007, 0705.0163.

[20]  Calibrating the nonlinear matter power spectrum: Requirements for future weak lensing surveys , 2004, astro-ph/0412142.

[21]  Fast Parameter Estimation from the CMB Power Spectrum , 2001, astro-ph/0108315.

[22]  A. Taylor,et al.  Cosmic microwave background temperature and polarization pseudo-Cl estimators and covariances , 2004, astro-ph/0410394.

[23]  A. Hamilton,et al.  Information content of the non-linear matter power spectrum , 2005, astro-ph/0502081.

[24]  B. Efron.  Bootstrap Methods: Another Look at the Jackknife , 1979 .

[25]  Edward J. Wollack,et al.  First year Wilkinson Microwave Anisotropy Probe (WMAP) observations: Determination of cosmological parameters , 2003, astro-ph/0302209.

[26]  B. Efron.  The jackknife, the bootstrap, and other resampling plans , 1987 .

[27]  M. Takada,et al.  The impact of non‐Gaussian errors on weak lensing surveys , 2008, 0810.4170.

[28]  T. Kitching,et al.  Analytic methods for cosmological likelihoods , 2010, 1003.1136.

[29]  Andrew R. Liddle,et al.  Model selection in cosmology , 2006 .

[30]  I. Szapudi,et al.  Shrinkage estimation of the power spectrum covariance matrix , 2007, 0711.2509.

[31]  Wayne Hu,et al.  � 1999. The American Astronomical Society. All rights reserved. Printed in U.S.A. POWER SPECTRUM TOMOGRAPHY WITH WEAK LENSING , 1999 .

[32]  Constraining isocurvature perturbations with cosmic microwave background polarization. , 2000, Physical review letters.

[33]  Olivier Ledoit,et al.  Improved estimation of the covariance matrix of stock returns with an application to portfolio selection , 2003 .

[34]  C. Baugh,et al.  Statistical analysis of galaxy surveys – I. Robust error estimation for two-point clustering statistics , 2008, 0810.1885.

[35]  Antony Lewis,et al.  Properties and use of CMB power spectrum likelihoods , 2009, 0902.0674.

[36]  A spherical harmonic analysis of redshift space , 1994, astro-ph/9409027.

[37]  I. Szapudi,et al.  FAST GENERATION OF ENSEMBLES OF COSMOLOGICAL N-BODY SIMULATIONS VIA MODE RESAMPLING , 2011, 1103.2767.

[38]  M. Rowan-Robinson,et al.  Spherical Harmonic Analysis of the PSCz Galaxy Catalogue: Redshift distortions and the real-space power spectrum , 1999, astro-ph/9901351.

[39]  O. Lahav,et al.  Massive lossless data compression and multiple parameter estimation from galaxy spectra , 1999, astro-ph/9911102.

[40]  A. Heavens,et al.  Fast parameter estimation from the cosmic microwave background power spectrum , 2002 .

[41]  P. Schneider,et al.  Why your model parameter confidences might be too optimistic - unbiased estimation of the inverse covariance matrix , 2006, astro-ph/0608064.

[42]  P. Schneider,et al.  Cosmic shear covariance: the log-normal approximation , 2011, 1105.3980.

[43]  Wendy L. Freedman,et al.  Report of the Dark Energy Task Force , 2006, astro-ph/0609591.

[44]  N. University,et al.  On measuring the covariance matrix of the non-linear power spectrum from simulations , 2005, astro-ph/0511416.

[45]  A. Heavens,et al.  Redshift-space distortions in the PSCz galaxy catalogue , 2001 .

[46]  M. Chevallier,et al.  ACCELERATING UNIVERSES WITH SCALING DARK MATTER , 2000, gr-qc/0009008.

[47]  Anton M. Koekemoer,et al.  A DETECTION OF WEAK-LENSING MAGNIFICATION USING GALAXY SIZES AND MAGNITUDES , 2011, 1111.3679.

[48]  Power Spectrum Covariance of Weak Gravitational Lensing , 2000, astro-ph/0012087.

[49]  Sho Matsumoto.  General Moments of the Inverse Real Wishart Distribution and Orthogonal Weingarten Functions , 2010, 1004.4717.

[50]  Masahiro Takada,et al.  Systematic errors in future weak-lensing surveys: requirements and prospects for self-calibration , 2006 .

[51]  S. Dye,et al.  The shear power spectrum from the COMBO-17 survey , 2003 .

[52]  U. Colorado,et al.  Information content of the non‐linear power spectrum: the effect of beat‐coupling to large scales , 2005, astro-ph/0511418.

[53]  N. Yoshida,et al.  SIMULATIONS OF BARYON ACOUSTIC OSCILLATIONS. II. COVARIANCE MATRIX OF THE MATTER POWER SPECTRUM , 2009, 0902.0371.

[54]  Nick Kaiser,et al.  Weak gravitational lensing of distant galaxies , 1992 .

[55]  M. Viel,et al.  Massive neutrinos and the non‐linear matter power spectrum , 2011, 1109.4416.

[56]  H. K. Eriksen,et al.  ESTIMATION OF POLARIZED POWER SPECTRA BY GIBBS SAMPLING , 2007 .

[57]  R. B. Barreiro,et al.  Size magnification as a complement to cosmic shear , 2012, 1209.1646.

[58]  Masanori Sato,et al.  SIMULATIONS OF WIDE-FIELD WEAK LENSING SURVEYS. I. BASIC STATISTICS AND NON-GAUSSIAN EFFECTS , 2009, 0906.2237.

[59]  Roberto Scaramella,et al.  Cosmology and Fundamental Physics with the Euclid Satellite , 2012, Living reviews in relativity.

[60]  M. Halpern,et al.  First-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Parameter Estimation Methodology , 2003 .

[61]  E. Linder.  Exploring the expansion history of the universe. , 2002, Physical review letters.

[62]  W. Kinney.  Constraining Inflation with Cosmic Microwave Background Polarization , 1998, astro-ph/9806259.

[63]  3D weak lensing , 2003, astro-ph/0304151.

[64]  R. Trotta.  Applications of Bayesian model selection to cosmological parameters , 2005, astro-ph/0504022.

[65]  M. Takada,et al.  Information content of weak lensing power spectrum and bispectrum: including the non-Gaussian error covariance matrix , 2012, 1207.6322.