| 作成者 |
|
|
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 終了ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| JaLC DOI |
|
| 関連DOI |
|
|
|
|
|
|
|
| 関連URI |
|
|
|
|
|
|
|
| 関連情報 |
|
|
|
|
|
|
|
| 概要 |
A nonparametric method of pattern recognition is proposed for directional observations, or for observations on a cyclic time series. Such observations, like birds migration, pollens' movement, bio-rhy...thm of living et al, can be rearranged directionally from an origin on a circumference of a unit circle. Then the authors suggest a possibility to apply a multiple circular rans test whether several sets of observations on a circle are drawn from the same distribution as a methodology of pattern recognition. Thus the present paper gives firstly the well-explained mathematical formulation of the generalized problem on the basis of combinatorial theory, secondly the numerical tables of critical numbers of runs with upper and lower significance levels and finally a compact FORTRAN program of proto-type for further cases of practical application of the present rnethod proposed.続きを見る
|
Mendeley出力
