Analysis of the long-run and one-way causality between macroeconomic index

宏观经济指标之间的长期单向因果关系分析

基本信息

批准号：
15530143
负责人：
YAO Feng
金额：
$ 1.73万
依托单位：
Kagawa University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2003
资助国家：
日本
起止时间：
2003 至 2005
项目状态：
已结题

项目摘要

In term of the project, we get the following results.1.Succeeded in transferred the program code written by the author at supper computer SX4 to personal computer (PC). Based on PC, we developed some programs by witch the calculation of one-way causal measures can be easily conducted.2.Based on cointegrated time series model, we showed the calculation algorithm of the one-way effect measures at time domain and at frequency domain. The proposed new method can contributed to reduce the cost of simulation.3.The calculation algorithms of long-run and short-run (local) one-way effects are made clearly. In addition, we developed the necessary calculation program based on Pro Fortran. By the use of these programs, it is possible to investigate the complex causal relationships in details.4.To apply the approach of one-way effect causal measures to the analysis of high frequency financial time series based on PC, the algorithm is made cleared5.To make use of the timely data published at internet, especially Chinese financial and market data, the way of transforming the date base file to text format is made cleared. By this way, we renew our time series data base of America and Japan as well as China.6.To conduct the project, we reported our research results at internal and international conferences and also have had made oversea researches.7.In this three years period, we published 5 research papers and reported 4 times at conference. Some of the new results will be reported at the 2006 Far Eastern Meeting of the Econometric Society which will be held in Tsinghua University, Beijing, China, on July 9-12, 2006.

就项目而言，我们将获得以下结果。1。在Supper Computer SX4编写的程序代码中取得了成功。基于PC，我们通过巫婆开发了一些程序，可以轻松地进行单向因果量度的计算。2。基于协调时间序列模型，我们显示了时间域和频域上单向效应测量的计算算法。提出的新方法可以降低仿真成本。3。长期和短期（局部）单向效应的计算算法清楚。此外，我们开发了基于Pro Fortran的必要计算程序。通过使用这些程序，可以详细研究复杂的因果关系。4。要应用单向效应因果关系措施的方法，以基于PC的高频财务时间序列的分析，对算法进行了清除。要清除算法，以利用在互联网上发布的及时数据，尤其是中国财务和市场数据，尤其是中国财务数据，将日期基础文件转换为文本格式的日期基础文件。通过这种方式，我们续签了美国和日本以及中国的时间序列数据库。6。为了进行该项目，我们在内部和国际会议上报告了我们的研究结果，并且已经进行了海外研究。7。在这三年中，我们发表了5篇研究论文并在会议上报告了4次。一些新的结果将在2006年的计量经济学学会远东会议上进行报道，该会议将于2006年7月9日至12日在中国北京的Tsinghua大学举行。