報(bào) 告 人：金百鎖 教授

報(bào)告題目：Spatial weights matrix selection and model averaging for spatial generalized linear model  

報(bào)告時(shí)間：2024年5月17日（周五）下午16:00

報(bào)告地點(diǎn)：靜遠(yuǎn)樓1508學(xué)術(shù)報(bào)告廳

主辦單位：數(shù)學(xué)與統(tǒng)計(jì)學(xué)院、數(shù)學(xué)研究院、科學(xué)技術(shù)研究院

報(bào)告人簡(jiǎn)介：

       金百鎖，中國(guó)科學(xué)技術(shù)大學(xué)管理學(xué)院統(tǒng)計(jì)與金融系教授。2001年畢業(yè)于中國(guó)科學(xué)技術(shù)大學(xué)獲得學(xué)士學(xué)位，2006年獲得中國(guó)科學(xué)技術(shù)大學(xué)博士學(xué)位。研究方向?yàn)樽兘Y(jié)構(gòu)模型，隨機(jī)矩陣，空間統(tǒng)計(jì)等。在PNAS，AoS，Biometrika，AAP等期刊已發(fā)表學(xué)術(shù)論文50余篇。主持承擔(dān)國(guó)家自然科學(xué)基金面上項(xiàng)目、重大項(xiàng)目課題、重點(diǎn)項(xiàng)目課題、國(guó)際交流項(xiàng)目等。現(xiàn)為中國(guó)現(xiàn)場(chǎng)統(tǒng)計(jì)研究會(huì)理事，中國(guó)現(xiàn)場(chǎng)統(tǒng)計(jì)研究會(huì)教育統(tǒng)計(jì)分會(huì)常務(wù)理事、秘書(shū)長(zhǎng)，中國(guó)現(xiàn)場(chǎng)統(tǒng)計(jì)研究會(huì)旅游大數(shù)據(jù)分會(huì)常務(wù)理事、副理事長(zhǎng)，中國(guó)現(xiàn)場(chǎng)統(tǒng)計(jì)研究會(huì)多元統(tǒng)計(jì)應(yīng)用專(zhuān)業(yè)委員會(huì) 常務(wù)理事。

報(bào)告摘要：

       Spatial weights matrix selection and model averaging for spatial generalized linear model. For analyzing non-normal data that are observed from all the spatial units, we proposed a generalized linear model, whose link function has autoregressive construction for spatial interaction. In this article, we develop an approach that uses instrumental variables (IVs) to derive maximum likelihood estimators for the parameters. It is conceptually simple and easier to implement. Under mild conditions, it is shown that the estimator resulting from two-stage MLE is consistent and asymptotically normally distributed. Base on the proposed method, we employ the Kullback-Leibler (KL) loss function with a penalty term (Zhang et al.,2016) to choose the true spatial weights matrix or the best one in the sense of minimizing of KL loss. Additionally, we introduce a model averaging procedure to effectively reduce the KL loss. Extensive simulation studies and data examples demonstrate the effectiveness of the proposed method.