POWER APPROXIMATIONS OF THE TEST OF HOMOGENEITY FOR MULTINOMIAL POPULATIONS
スポンサーリンク
概要
- 論文の詳細を見る
In this paper we propose three approximations to the power of the power-divergence statistic Ra on the test of homogeneity for multinomial populations. These approximations are based on the approximation theories for the multinomial goodness-of-fit test as proposed by Broffitt and Randles (1977) and Drost et al. (1989). One approximation is constructed from a limiting normal distribution of Ra. The others are constructed from the linear and quadratic terms of a Taylor series expansion of Ra. These approximations are compared with the power directly calculated from the product multinomial model. As a result of this comparison, the latter approximations are found to perform well when a≥0. On the homogeneity of two populations, we investigate the power of representative power divergence statistics R0, R2/8 and R1 using the latter approximations. Then, corresponding to alternatives, relations to the power of the three statistics are discussed.
- 日本統計学会の論文
日本統計学会 | 論文
- 高次元多変量分析におけるパーミュテーションテスト(日本統計学会75周年記念特集(II))
- E-2 Wishart行列の固有ベクトルに関するパーミュテーション検定について
- D-1 相関係数に関するいくつかの検定問題について(多変量解析(1))(日本統計学会第69回大会記録)
- A-1 ベイズ情報量基準のある適用における妥当性(日本統計学会第67回大会記録 : 統計一般理論(3)ベイズ統計学)
- 教育・学習支援のためのデータ指向統計解析環境(統計教育)