Multivariate Non-Normality in the WMAP 1st Year Data

Details Multivariate Non-Normality in the WMAP 1st Year Data Multivariate Non-Normality in the WMAP 1st Year Data

2005-11-29

arxiv.org/abs/astro-ph/0511802v1

Astronomy and Astrophysics

Astrophysics

15 pages, 24 figures, submitted to MNRAS

Scientific paper

The extraction of cosmological parameters from microwave background observations relies on specific assumptions about the statistical properties of the data, in particular that the p-point distributions of temperature fluctuations are jointly-normal. Using a battery of statistical tests, we assess the multivariate Gaussian nature of the Wilkinson Microwave Anisotropy Probe (WMAP) 1st year data. The statistics we use fall into three classes which test different aspects of joint-normality: the first set assess the normality of marginal (one-point) distributions using familiar univariate methods; the second involves statistics that directly assess joint-normality; and the third explores the evidence of non-linearity in the relationship between variates. We applied these tests to frequency maps, `foreground-cleaned' assembly maps and all-sky CMB-only maps. The assembly maps are of particular interest as when combined with the kp2 mask, we recreate the region used in the computation of the angular power spectrum. Significant departures from normality were found in all the maps. In particular, the kurtosis coefficient, D'Agostino's statistic and bivariate kurtosis calculated from temperature pairs extracted from all the assembly maps were found to be non-normal at 99% confidence level. We found that the results were unaffected by the size of the Galactic cut and were evident on either hemisphere of the CMB sky. The latter suggests that the non-Gaussianity is not simply related to previous claims of north-south asymmetry or localized abnormalities detected through wavelet techniques.

Affiliated with

Also associated with

No associations

Rating

Multivariate Non-Normality in the WMAP 1st Year Data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Multivariate Non-Normality in the WMAP 1st Year Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multivariate Non-Normality in the WMAP 1st Year Data will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-715437