TechRxiv
Consistency and Asymptotic Normality of the Maximum Likelihood Estimator in GaGLM.pdf (222.8 kB)

Consistency and Asymptotic Normality of the Maximum Likelihood Estimator in GaGLM

Download (222.8 kB)
preprint
posted on 22.02.2021, 22:31 by Benchao Wang, Hong Gu, Pan Qin
This paper theoretically investigates the asymptotic properties of maximum likelihood estimates of GaGLM, and discusses some properties about Gamma distribution. It can provide theoretical foundation for expanding the application scope of gamma distribution based regression model, and benefit the further interval estimates, hypothesis tests and stochastic control design.The existence of the Gamma function in Gamma distribution makes correlation analysis certain specificity, and few researchers do relevant theoretical research. To complement this part, we established the asymptotic properties and the application condition of maximum likelihood estimates of GaGLM. In addition to this, we also discussed the propertis of the Fisher information matrix ,and n-order moment of Z and log(Z).

Funding

National Natural Science Foundation of China 61502074

National Natural Science Foundation of China 61633006

National Natural Science Foundation of China U1560102

History

Email Address of Submitting Author

wangbc@mail.dlut.edu.cn

Submitting Author's Institution

Dalian University of Technology

Submitting Author's Country

China

Licence

Exports