Cosmological Constraints from the Cosmic Microwave Background

TL;DR

This paper analyzes first-generation cosmic microwave background data to constrain inflationary cosmological parameters, confirming simple inflation models and highlighting the importance of baryon content and calibration improvements.

Contribution

It introduces an approximate likelihood method for band-power estimates and provides new constraints on cosmological parameters from CMB data.

Findings

Flat geometries and scale-invariant spectra are favored.

Models with significant negative curvature are eliminated.

High baryon content is indicated by the data.

Abstract

Using an approximate likelihood method adapted to band--power estimates, we analyze the ensemble of first generation cosmic microwave background anisotropy experiments to deduce constraints over a six--dimensional parameter space describing Inflation--generated adiabatic, scalar fluctuations. The basic preferences of simple Inflation scenarios are consistent with the data set: flat geometries $(\OmT \equiv 1-\Omk \sim 1)$ and a scale--invariant primeval spectrum ($n\sim 1$) are favored. Models with significant negative curvature ($\OmT < 0.7$) are eliminated, while constraints on postive curvature are less stringent. Degeneracies among the parameters prevent independent determinations of the matter density $\OmM$ and the cosmological constant $\Lambda$, and the Hubble constant $\Ho$ remains relatively unconstrained. We also find that the height of the first Doppler peak relative to the amplitude suggested by data at larger $l$ indicates a high baryon content ($\Omb h^2$), almost independently of the other parameters. Besides the overall qualitative advance expected of the next generation experiments, their improved dipole calibrations will be particularly useful for constraining the peak height. Our analysis includes a {\em Goodness--of--Fit} statistic applicable to power estimates and which indicates that the maximum likelihood model provides an acceptable fit to the data set.