主題:模型選擇原理和多模型統(tǒng)計(jì)推斷
主講人:喻達(dá)磊博士
時(shí)間:2015年3月20日(周五)9:20
地點(diǎn):北院卓遠(yuǎn)樓305
主辦單位:統(tǒng)計(jì)與數(shù)學(xué)學(xué)院
摘要:
Model selection is essentially important when there are different statistical models at hand. In this talk, we consider two important model selection criteria, i.e. Akaike information criterion (AIC) and Bayesian information criterion (BIC). The derivation of these model selection criteria will be introduced and the underlying principles will be discussed. Then the notion of AIC will be extended into the domain of mixed effects models and some new results will be reported.
主講人喻達(dá)磊簡介:
香港城市大學(xué)統(tǒng)計(jì)學(xué)博士,,副教授。研究領(lǐng)域?yàn)殡S機(jī)效應(yīng)模型,、混合模型和空間計(jì)量經(jīng)濟(jì)學(xué)模型的統(tǒng)計(jì)推斷,。在Stat. Med.、J. Multivariate Anal.,、Comput. Stat. Data Ana.等國際統(tǒng)計(jì)期刊發(fā)表論文多篇,,擔(dān)任過Comput. Stat. Data Ana.和Hydrological Sci. J. 等國際期刊的匿名審稿人。
English version
Bi-weekly Research Seminar Series
Topic Model selection principles and multi-model inference
Speaker
Dr. Dalei Yu
Associate Professor
Department of Statistics
Yunnan University of Finance and Economics
Date 20 March 2015
Time 9:20 – 11:20 a.m.
Venue 305 Zhuo Yuan Bd.
Abstract:
Model selection is essentially important when there are different statistical models at hand. In this talk, we consider two important model selection criteria, i.e. Akaike information criterion (AIC) and Bayesian information criterion (BIC). The derivation of these model selection criteria will be introduced and the underlying principles will be discussed. Then the notion of AIC will be extended into the domain of mixed effects models and some new results will be reported.
All are welcome!