JaLCDOI 10.18926/fest/52227
Title Alternative Numerical Evaluation of the Approximation by an Influence Function
FullText URL fest_019_001_007.pdf
Author Bao, Hong Mei| Fueda, Kaoru|
Abstract For evaluating statistical models one of the most effective criteria is cross-validation. But it requires a large amount of computation. Various alternative schemes are considered to reduce its computation. Modified generalized information criterion is one of those alternative schemes. In this criterion an influence function is used to estimate the parameters of the models. By the numerical simulation we studied the effect of an influence function.  Surveying data of the lake depth are used as the sample data. We estimate the shape of lake bottom as spline surface. The estimated parameters and the estimated depths obtained by two criteria are compared and the effect of an information function is analysed.
Keywords influence function information criteria CV mGIC B-spline
Publication Title 岡山大学環境理工学部研究報告
Published Date 2014-03
Volume volume19
Issue issue1
Start Page 1
End Page 7
ISSN 2187-6940
language 英語
File Version publisher
NAID 120005394642
JaLCDOI 10.18926/fest/44790
Title Alternative Validation of Dollar cost averaging investment method
FullText URL fest_016_001_005.pdf
Author Touji, Masayuki| Fueda, Kaoru|
Abstract As a method of long term investment for private investor, the dollar cost averaging investment method is well known and seems to reduce the purchase cost because we purchase risk assets with same amount of monye every month, then we purchase many assets when the price of assets is low and few assets when the price is high. On the other hand, if the expectation of the return of the risk assets is positive, we have the maximum expectation of return when we purchase the risk assets with all of money to invest. To reduce the risk of investment, diversified investments are effective. However question whehter we use the dollar cost averaging investment method or investe money all at once to well-diversified risk assets remains. In this study, we validate the effecte of the ddollar cost averaging investment method by Monte Carlo simulation.
Keywords Dollar cost averaging investment Monte Carlo simulation Stock price index
Publication Title 岡山大学環境理工学部研究報告
Published Date 2011-03-18
Volume volume16
Issue issue1
Start Page 1
End Page 5
ISSN 2185-3347
language 日本語
File Version publisher
NAID 120002949470
JaLCDOI 10.18926/fest/12816
Title Alternative Statistical modeling for analysis of malaria epidemic behavior at Ishigaki Island
FullText URL 13_007_015.pdf
Author Ueki, Masao| Nakagawa, Yuuki| Fueda, Kaoru| Ishikawa, Hirofumi|
Abstract It is necessary to consider a stochastic variability in modeling malaria epidemic behavior since the malaria infection cycle essentially depends on stochastic elements. For this requirement, we need to construct an appropriate statistical model from available data in advance. In this report, we provide some statistical models for the analysis of malaria epidemic behavior at Ishigaki Island. These models can be used for recurrence of past malaria epidemic and prediction of future malaria epidemic at Ishigaki Island.
Keywords Anopheles minimus, generalized liner model Ishigaki Island malaria epidemic behavior prediction stochastic model
Publication Title 岡山大学環境理工学部研究報告
Published Date 2008-03
Volume volume13
Issue issue1
Start Page 7
End Page 15
ISSN 1341-9099
language 日本語
File Version publisher
NAID 120002304876
JaLCDOI 10.18926/fest/11429
Title Alternative Variable selection besed on global score estimation and its numerical investigation
FullText URL 012_029_040.pdf
Author Mori, Yuichi| Fueda, Kaoru| Iizuka, Masaya|
Abstract A variable selection method using global score estimation is proposed, which is applicable as a selection criterion in any multivariate method without external variables such as principal component analysis. This method selects a reasonable subset of variables so that the global scores, e.g. principal component scores, which are computed based on the selected variables, approximate the original global scores as well as possible in the context of the least squares. Three computational steps are proposed to estimate the scores according to how to satisfy the restriction that the estimated global scores are mutually uncorrelated. Three different examples are analyzed to demonstrate the performance and usefulness of the proposed method numerically, in which three steps are evaluated and the results obtained using four cost-saving selection procedures are compared.
Keywords principal components least square orthogonalization cost-saving selection
Publication Title 岡山大学環境理工学部研究報告
Published Date 2007-03-15
Volume volume12
Issue issue1
Start Page 29
End Page 40
ISSN 1341-9099
language 日本語
File Version publisher
NAID 120002313992
JaLCDOI 10.18926/fest/11428
Title Alternative Socially Responsible Investment to Japanese Companies
FullText URL 011_031_041.pdf
Author Ikeda, Yousuke| Fueda, Kaoru|
Abstract Recently, Socially Responsible Investment, which is a policy of investment regarding with companies' social, environmental and moral value, attracts attention of investors. However, there are little data to explain the effectiveness of the SIR in Japan. In this paper we report the relation between environmental management and stock price of Japanese companies.
Keywords Socially Responsible Investment Environmental rating Eco fund
Publication Title 岡山大学環境理工学部研究報告
Published Date 2006-03-15
Volume volume11
Issue issue1
Start Page 31
End Page 41
ISSN 1341-9099
language 日本語
File Version publisher
NAID 120002313974
JaLCDOI 10.18926/fest/11427
Title Alternative Kernel density estimation on the interval
FullText URL 011_027_030.pdf
Author Ueki, Masao| Fueda, Kaoru|
Abstract In the field of data analysis, including environmental data, it is important to know the shape of underlying density function. In this case, we often use histogram which provides an information about the board line of density's curve. However histogram can not be the best method when the true density function is continuous, as is often the cases. On the other hand, kernel density estimator is another popular one which gives a continuous function. In some practical cases, however, there is a case that some knowledges about the range of the data are previously given. For instance, data of percentage, such as mortality rate, only takes the values on [0,1]. This paper considers two different modifications in kernel density estimator for the data on known interval and compares them.
Keywords kernel density estimator adaptive bandwidth data on finite interval
Publication Title 岡山大学環境理工学部研究報告
Published Date 2006-03-15
Volume volume11
Issue issue1
Start Page 27
End Page 30
ISSN 1341-9099
language 日本語
File Version publisher
NAID 120002313832