dainty
发表于 2025-3-25 07:16:20
http://reply.papertrans.cn/88/8766/876568/876568_21.png
运动的我
发表于 2025-3-25 09:56:14
http://reply.papertrans.cn/88/8766/876568/876568_22.png
–LOUS
发表于 2025-3-25 14:33:23
Score Tests in Overdispersed GLM’siance function. Two versions of the tests on means are considered, one that assumes the specified mean-variance relation to be correct and another that has a more general asymptotic justification. Applications are made to log-linear models for overdispersed Poisson data with negative binomial variance function.
insurrection
发表于 2025-3-25 16:07:57
Estimation and Tests in a Quasi-likelihood Model with a Non-constant Dispersion Parameterrs. The algorithm is derived and may be interpreted as an iterative weighted least squares method, which combines nicely with the structure of GLIM. The algorithm is applied to an example based on double logistic fit. Further tests on the linear parameters are discussed.
伪造
发表于 2025-3-25 22:05:49
http://reply.papertrans.cn/88/8766/876568/876568_25.png
字的误用
发表于 2025-3-26 01:32:03
Modelling Transition Probabilities in the Analysis of Aggregated Dataors and covariates on overdispersion (as well as on transition probabilities) can be investi-gated in a flexible manner. An application to Italian electoral data is discussed in some detail and the main features of a maximization routine based on the Fisher scoring algorithm are also outlined.
adj忧郁的
发表于 2025-3-26 04:48:08
http://reply.papertrans.cn/88/8766/876568/876568_27.png
Mumble
发表于 2025-3-26 08:45:05
Functional programming for GLMs account is given of functional programming concepts. The prototype language used (Standard ML) is discussed and some of its strengths and weaknesses outlined. Some ideas are presented on how this approach might be modified to give a modern computing environment for a statistical modelling package of the future.
Anticonvulsants
发表于 2025-3-26 13:07:41
http://reply.papertrans.cn/88/8766/876568/876568_29.png
斗争
发表于 2025-3-26 20:30:49
A Few Problems with Application of the Kalman Filter practice, some problems have to be solved before confidently using the Kalman filter. These problems are related both with the numerical accuracy of the algorithm proposed by Kalman, and with the estimation of parameters that in the conventional Kalman filter are assumed to be known..