Context. Several authors have claimed to detect asignificant cross-correlation between microwave WMAP anisotropies andthe SDSS galaxy distribution. We repeat these analyses to determine thedifferent cross-correlation uncertainties caused by re-sampling errorsand field-to-field fluctuations. The first type of error concernsoverlapping sky regions, while the second type concerns non-overlappingsky regions. Aims. To measure the re-sampling errors, we use bootstrap andjack-knife techniques. For the field-to-field fluctuations, we usethree methods: 1) evaluation of the dispersion in the cross-correlationwhen correlating separated regions of WMAP with the original region ofSDSS; 2) use of mock Monte Carlo WMAP maps; 3) a new method (developedin this article), which measures the error as a function of theintegral of the product of the self-correlations for each map. Methods. The average cross-correlation for b>30degis significantly stronger than the re-sampling errors - both thejack-knife and bootstrap techniques provide similar results - but it isof the order of the field-to-field fluctuations. This is confirmed bythe cross-correlation between anisotropies and galaxies in more thanthe half of the sample being null within re-sampling errors. Results. Re-sampling methods underestimate the errors.Field-to-field fluctuations dominate the detected signals. The ratio ofsignal to re-sampling errors is larger than unity in a way thatstrongly depends on the selected sky region. We therefore conclude thatthere is no evidence yet of a significant detection of the integratedSachs-Wolfe (ISW) effect. Hence, the value of obtained by the authors who assumed they were observing the ISW effectwould appear to have originated from noise analysis. Key words: cosmic microwave background - large-scale structure of Universe
展开▼
