Computing Environments for Stastistics

研究者探索和发展新的原则，统计软件的设计，以利用现代计算能力。特别强调的是探索统计语言的并行计算、编译和代码分析的有效使用。试点实施被纳入开源统计软件系统。通过统计软件提供的有效统计方法对于研究人员最大限度地利用实验和观测数据的能力以及教师教授良好研究实践的能力至关重要。本研究开发的设计原则导致软件提高了研究人员，教师和统计方法的其他用户在科学研究和教学中更有效地应用该方法的能力，并充分利用现代高性能计算资源。这些原则还导致了软件框架，可以用来更快地向最终用户交付新的统计方法。在一系列领域的应用为本研究中开发的方法和原理提供了测试平台。

The investigator explores and develops new principles for the designof statistical software to take advantage of modern computing power.Particular emphasis is placed on exploring the effective use ofparallel computing, compilation, and code analysis for statisticallanguages. Pilot implementations are incorporated in open sourcestatistical software systems.Effective statistical methodology made available through statisticalsoftware is critical to the ability of researchers to make maximal useof experimental and observational data, and for the ability ofinstructors to teach good research practices. The design principlesdeveloped by this research lead to software that improves the abilityof researchers, instructors, and other users of statisticalmethodology to apply this methodology more effectively in scientificresearch and teaching and to take full advantage of modernhigh-performance computational resources. These principles also leadto software frameworks that can be used to more rapidly deliver newstatistical methodology to end users. Applications in a range ofareas serve as testbeds for methods and principles developed in thisresearch.