In this article, we proposed the diagonstic statistic for evaluating influence of observations on parameter vector in nonlinear regression. This statistic, D₁^(*)(H^, qσ²), is a modified formula for Cook's statistic, D₁(Z^/Z^, qσ²), s...
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
https://www.riss.kr/link?id=A40002333
1992
Korean
304.000
학술저널
39-48(10쪽)
0
상세조회0
다운로드다국어 초록 (Multilingual Abstract)
In this article, we proposed the diagonstic statistic for evaluating influence of observations on parameter vector in nonlinear regression. This statistic, D₁^(*)(H^, qσ²), is a modified formula for Cook's statistic, D₁(Z^/Z^, qσ²), s...
In this article, we proposed the diagonstic statistic for evaluating influence of observations on parameter vector in nonlinear regression.
This statistic, D₁^(*)(H^, qσ²), is a modified formula for Cook's statistic, D₁(Z^/Z^, qσ²), so that it includes the components of the curvature on the nonlinear curve.
Hence, D₁^(*) (H^, qσ²) is able to be used to identify the observations located near on seriously curved position as well as high leveraged points or outliers.
노동생산성 향상을 위한 회계시스템이 기업성과에 미치는 영향에 관한 연구