2005 Fiscal Year Final Research Report Summary
Research for the constructions of designs in the design of experiments
| Project/Area Number |
16540114
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
General mathematics (including Probability theory/Statistical mathematics)
|
|---|
| Research Institution | Osaka Prefecture University |
|---|
Principal Investigator |
KURIKI Shinji Osaka Prefecture University, Graduate School of Engineering, Professor, 工学研究科, 教授 (00167389)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ARIZONO Ikuo Osaka Prefecture University, Graduate School of Engineering, Associate Professor, 工学研究科, 助教授 (20175988)
SHIRAKURA Teruhiro Kobe University, Faculty of Human Development, Professor, 発達科学部, 教授 (30033913)
|
|---|
| Project Period (FY) |
2004 – 2005
|
| Keywords | The design of experiments / Constructions / Split-block designs / Split-plot designs |
|---|
| Research Abstract |
The designs considered under this research are split-block designs and split-plot designs, which are two special types of the factorial experiments with two factors. Hering and Mejza, S. (1997,2002) etc. gave the constructions of split designs using the Kronecker product of two subdesigns. But the split designs constructed by them are larger and the designs are not useful practically. So we were planning to construct smaller split designs. First of all, we defined the partial Kronecker product of two subdesigns. We constructed split-block designs and we characterized the statistical properties of the split-block designs, which is seen in Mejza, I., Kuriki and Mejza, S. (2001).
In 2004, we considered split-block designs and we generalized the results given by Mejza, I., Kuriki and Mejza, S. (2001). Here we generally obtained the standard contrasts of the treatment parameters in the multistratum analysis. This result is very interesting, because we can usually obtain the stratum efficiency factors, but we cannot obtain the standard contrasts in the general form.
In 2005, we considered split-plot designs. We constructed smaller split-plot designs and we characterized the statistical properties of the split-plot designs by use of the results obtained in 2004.
As the next theme of our research, we are planning to construct split designs with control treatments and test treatments. For example, control treatments are the standard varieties of corn and test treatments are new varieties in the agricultural experiments. We expect that the results for the split designs in 2004 and 2005 will be applied to construct such split designs.
|
Research Products
(39 results)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Book] 数理統計学2006
Author(s)
長尾壽夫
Total Pages
242
Publisher
共立出版社
Description
「研究成果報告書概要(和文)」より
