RUI: New Parameter Estimation, Data Analysis and Statistical Strategies for LIGO

RUI：LIGO 的新参数估计、数据分析和统计策略

• 批准号: 0071327
• 负责人: Nelson Christensen
• 金额: $ 9.97万
• 依托单位: Carleton College
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2000
• 资助国家: 美国
• 起止时间: 2000-07-15 至 2003-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0071327&HistoricalAwards=false
• 关键词: RUI; New; Parameter; Estimation; Data

The principal investigator will conduct research with the Laser Interferometer Gravitational-Wave Observatory (LIGO). The effort will be based at Carlton College, an undergraduate institution. The LIGO construction is nearly complete and the facility is being commissioned for operations. A first science run is scheduled for 2002. The focus of this research will be the analysis of data from LIGO.The project involves a new strategy for the statistical analysis of data recorded by LIGO. Over the last decade there has been a revolution in the computational methods for treating Bayesian statistics. Without these new methods the multidimensional integrals over the parameter space make a Bayesian analysis computationally prohibitive. With the new technique known as Markov Chain Monte Carlo the integrals are evaluated by a controlled random walk through the parameter space. This research will apply the techniques of Markov Chain Monte Carlo to LIGO data in an attempt to identify the signature of gravitational radiation and to estimate the parameters of the astrophysical source of the gravitational radiation.

首席研究员将与激光干涉仪引力波天文台(LIGO)进行研究。这项工作将在本科生机构卡尔顿学院进行。LIGO的建设接近完成，该设施正在投入运营。第一次科学运行计划于2002年进行。这项研究的重点将是对LIGO的数据进行分析。该项目涉及到对LIGO记录的数据进行统计分析的新策略。在过去的十年里，处理贝叶斯统计的计算方法发生了一场革命。如果没有这些新的方法，参数空间上的多维积分使得贝叶斯分析在计算上是不可能的。利用新的马尔可夫链蒙特卡罗技术，通过在参数空间的受控随机游动来计算积分。这项研究将把马尔科夫链蒙特卡罗技术应用于LIGO数据，试图识别引力辐射的特征，并估计引力辐射的天体物理源的参数。