Advances and Applications in Statistics
Volume 34, Issue 1, Pages 49 - 63
(May 2013)
|
|
ON THE GEOMETRY OF THE PARAMETER SPACE REDUCTION WITH APPLICATION TO GRAVITATIONAL WAVE DATA ANALYSIS
Robert J. Budzyński, Witold Kondracki and Andrzej Królak
|
Abstract: We study the problem of estimation of parameters in statistical inference. We propose a geometrical construction defining the reduction of the parameter space that can be used in estimation of parameters using maximum likelihood approach. We define precise conditions and assumption under which such a reduction exists. The reduction of the size of the parameter space can diminish enormous computational cost that is usually needed to search for gravitational wave signals and it can eliminate nuisance parameters that are not interesting in the analysis. We illustrate our proposal with several examples that have application to estimation of parameters of the signals in noise, in particular to estimation of parameters of the gravitational waves signals. |
Keywords and phrases: parameter estimation, geometric inference, gravitational wave data analysis. |
|
Number of Downloads: 370 | Number of Views: 1322 |
|