Statistical analysis system for supporting simulation and resampling

支持模拟和重采样的统计分析系统

• 批准号: 11480056
• 负责人: YAMAMOTO Yoshikazu
• 金额: $ 7.17万
• 依托单位: Tokushima Bunri University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 1999
• 资助国家: 日本
• 起止时间: 1999 至 2002
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-11480056/
• 关键词: Distributed processing; Knowledge representation; Statistical system; Java language; 並列分散処理; 支援システム; Java Spaces; Jini; XploRe; Bootstrap; グラフィカルユーザインタフェース; オブジェクト指向; 統計モデル

The aim of this research is to build a statistical system which can support simulation and resampling methods with easy and simple operations through an advanced user interface and a useful programming language. Simulation and resampling methods are computer intensive statistical methods, and are especially useful for complicated statistical problems. However, few statistical analysis systems support these methods efficiently. Therefore, we have developed a prototype statistical system named Jasp (Java based Statistical System) that equips various effective features to support simulation and resampling methods.A Jasp language is a function-based programming language and has ability for object-oriented programming. Parallel and distributed computing abilities of Jasp can be easily used even by users who are not computer experts, and are especially useful for simulation and resampling methods to execute required huge amount of calculation by using several personal computer simultaneously by a simple program written in the Jasp language. The Jasp user interface, which has a graphical user interface and a character user interface, is designed for using the Jasp language easily. Jasp has a function to utilize statistical knowledge for analysis. Jasp has also flexible extensibility ; it is used, for example, to get a dataset in Microsoft Excel and to display graphs of other software packages.

本研究的目的是通过一个先进的用户界面和一种有用的编程语言，建立一个能够支持模拟和重抽样方法的统计系统，操作简单易懂。模拟和重抽样方法是计算机密集型统计方法，特别适用于复杂的统计问题。然而，很少有统计分析系统有效地支持这些方法。因此，我们开发了一个原型统计系统JASP(Java Based Statistics System)，它具有支持模拟和重抽样方法的各种有效功能。JASP语言是一种基于函数的编程语言，具有面向对象的编程能力。即使不是计算机专家的用户也可以使用JASP的并行和分布式计算能力，特别是对于模拟和重采样方法，通过用JASP语言编写的简单程序同时使用多台个人计算机来执行所需的大量计算。JASP用户界面具有图形用户界面和字符用户界面，旨在方便地使用JASP语言。JASSP具有利用统计知识进行分析的功能。Jasp还具有灵活的可扩展性；例如，它用于在Microsoft Excel中获取数据集，以及显示其他软件包的图形。

项目成果

Y. Yamamoto and J. Nakano: "Distributed computing in a time series analysis system"Journal of the Japanese Society of Computational Statistics. Vol.15, No.1. 65-77 (2002)

Y. Yamamoto 和 J. Nakano：《时间序列分析系统中的分布式计算》日本计算统计学会杂志。

I. Kobayashi et al.: "A Procedural and Object-Oriented Statistical Language"Computational Statistics. Vol.17, No.3. 395-410 (2002)

I. Kobayashi 等人：“一种过程和面向对象的统计语言”计算统计。

藤原丈史, 中野純司, 山本由和, 小林郁典: "Java言語による統計解析システムJasp"統計数理. 49・2. 333-343 (2001)

Takeshi Fujiwara、Junji Nakano、Yoshikazu Yamamoto、Ikunori Kobayashi：“使用 Java 语言的统计分析系统 Jasp” 统计数学 49・2。

Fujiwara T., Nakano J., Yamamoto Y., Kobayashi I.: "An implementation of a statistical language based on Java"Journal of the Japanese Society of Computational Statistics. 14・1(掲載予定). (2001)

Fujiwara T.、Nakano J.、Yamamoto Y.、Kobayashi I.：“基于 Java 的统计语言的实现”日本计算统计学会期刊 14・1（待出版）。

I. Kobayashi et al.: "Extensibilities of a java based statistical system"The Japanese Society of Computational Statistics. (To appear). (2002)

I. Kobayashi 等人：“基于 java 的统计系统的扩展性”日本计算统计学会。