Quantum version of the curved exponential family is studied where its dual structure with different metrics.
研究量子统计中曲指数族子流形的不同度量的对偶结构。
Q function have been used to study the statistical diagnostics for the exponential family nonlinear models with random effects.
对指数族非线性随机效应模型，本文首次利用Q函数进行统计诊断。主要创新性研究成果如下。
Especially, we give some results of the convergence of the EM algorithm for the curved exponential family under the conditions checked easily.
特别对应用广泛的曲指数族，本文在较易实际验证的条件下给出了相应的EM算法的收敛结果。
In the second we give the necessary and sufficient conditions for strict convexity of power function of exponential family in the discrete case.
在第二个问题中，我们给出了在离散的情形下，指数族势函数严凸的充分必要条件。
The generalized nonlinear models (GNLMs), i. e. exponential family nonlinear models are the natural generalization of ordinary nonlinear models.
广义非线性模型又称指数族非线性模型，它是普通正态非线性模型的推广。
In this paper, a Bayes decision rule is derived for the scale-exponential family with error in variables, and an empirical Bayes (EB) decision rule is constructed by a deconvolution kernel method.
本文获得了刻度指数族变量带误差情形下的贝叶斯决策，且利用解卷积的核方法构造出了经验贝叶斯决策。
This paper discusses the exponential regression model, which is the simplest but very important in the family of Bayesian accelerated failure-time model.
讨论了加速失效模型族中最简单而又十分重要的指数回归模型，利用贝叶斯方法提高了该模型的有效性。
Firstly, the optimality of identical sensor system for the exponential distribution family has been verified.
首先，我们确证了全同感测器系统在指数分布族的参数辨识问题上的最佳性。
Methods A simple method for distinguishing of exponential distribution family is obtained based on factorization theorem.
方法在因子分解定理的基础上提出了一个判别指数型分布族的简单方法。
Results a factorization discriminance of exponential distribution family was obtained, moreover good properties of exponential distribution family were established.
结果提出了指数型分布族的因子分解判别法，并给出了指数型分布族的一个良好性质。
Results a factorization discriminance of exponential distribution family was obtained, moreover good properties of exponential distribution family were established.
结果提出了指数型分布族的因子分解判别法，并给出了指数型分布族的一个良好性质。
应用推荐