In Annual Review of Nuclear and Particle Science, vol. 59, issue 1, pp. 95-114, 2009.
http://arxiv.org/abs/0903.4210

For a PDF version of the article, click here.

STATISTICAL METHODS FOR COSMOLOGICAL PARAMETER SELECTION AND ESTIMATION

Andrew R Liddle


Astronomy Centre, University of Sussex, Brighton BN1 9QH, UK


Abstract: The estimation of cosmological parameters from precision observables is an important industry with crucial ramifications for particle physics. This article discusses the statistical methods presently used in cosmological data analysis, highlighting the main assumptions and uncertainties. The topics covered are parameter estimation, model selection, multi-model inference, and experimental design, all primarily from a Bayesian perspective.


Table of Contents

INTRODUCTION

INFERENCE
Orientation
Bayesian inference
Alternatives to Bayesian inference

COSMOLOGICAL PARAMETER ESTIMATION
Goals and methodology
Monte Carlo methods
Uncertainties, biases and significance

COSMOLOGICAL MODEL SELECTION
Model selection versus parameter estimation
The Bayesian evidence
Calculational methods
Multi-model inference
Other approaches to model selection

FORECASTING AND EXPERIMENTAL DESIGN
Fisher matrix approaches
Model selection approaches

THE END

REFERENCES

Next