| SOME ASYMPTOTIC INFERENCE IN MULTINOMIAL NONLINEAR MODELS （A GEOMETRIC APPROACH） |
| |
| 摘 要： | ＳＯＭＥＡＳＹＭＰＴＯＴＩＣＩＮＦＥＲＥＮＣＥＩＮＭＵＬＴＩＮＯＭＩＡＬＮＯＮＬＩＮＥＡＲＭＯＤＥＬＳ（ＡＧＥＯＭＥＲＩＣＡＰＰＲＯＡＣＨ）￥ＷＥＩＢＯＣＨＥＮＧ（ＤｅｐａｒｔｍｅｎｔｏｆＭａｔｈｅｍａｔｉｃｓ，ＳｏｕｔｈｅａｓｔＵｎｉｖｅｒｓｉｔｙ...
|
| 关 键 词： | 渐进线推理 多项式非线性模型 几何逼近 几何构造 |
| 收稿时间： | 15 September 1994 |
Some asymptotic inference in multinomial nonlinear models (A geometric approach) |
| |
| Authors: | Wei Bocheng |
| |
| Institution: | (1) Department of Mathematics, Southeast University, 210096 Nanjing |
| |
| Abstract: | A geometric framework is proposed for multinomial nonlinear models based on a modified version of the geometric structure
presented by Bates & Watts 4]. We use this geometric framework to study some asymptotic inference in terms of curvatures
for multinomial nonlinear models. Our previous results 15] for ordinary nonlinear regression models are extended to multinomial
nonlinear models. |
| |
| Keywords: | Curvature array multinomial nonlinear models information loss observed information stochastic expansion variance |
| 本文献已被 SpringerLink 等数据库收录！ |
