創(chuàng)源大講堂學(xué)術(shù)講座系列

講座題目： WHY ARE SEMIPARAMETRIC MODELS USEFUL

報(bào)告人：Liang, Hua教授，美國(guó)喬治華盛頓大學(xué)

主持人：鄭海濤教授

時(shí)間： 6月27號(hào)，周五，下午：4:00pm-5:00pm；

地點(diǎn)： 犀浦校區(qū)二號(hào)樓kaiyun開(kāi)云官方網(wǎng)站 X2511

內(nèi)容簡(jiǎn)介：In this talk, I will present several semi-parametric models including partial linear additive models and generalized partial linear additive models, which can relax structure assumption for linear or generalized linear models, and briefly mention the associated methods we have developed. I will elucidate them by using several real datasets to show why semi-parametric models are helpful and useful in health science and biomedical research.  Examples include a nutritional epidemiology study for exploring the relationship between plasma beta-carotene levels and personal characteristics and dietary factors, a HIV/AIDS study of the relationship between the use  of condom  and other indexes such as CD4 cell count and marital status, whether diabetes disease is related to BMI and age. Finally I will show you how to explore the relationship between cancer and genes when the number of gene expressions is larger than the number of subjects.

報(bào)告人簡(jiǎn)介：Liang， Hua，博士，美國(guó)喬治華盛頓大學(xué)統(tǒng)計(jì)學(xué)教授、博士生導(dǎo)師。現(xiàn)主要從事偏線性統(tǒng)計(jì)、變量選擇，混合模型，半?yún)?shù)回歸，臨床試驗(yàn)等方面的研究工作。在國(guó)外頂級(jí)學(xué)術(shù)期刊上發(fā)表論文二十余篇。主持美國(guó)國(guó)家自然科學(xué)基金、NIH基金項(xiàng)目等項(xiàng)目七項(xiàng)。目前為SCI期刊JASA， Electronic Journal of Statistics 副主編。Fellow of the American Statistical Association，the Institute of Mathematical Statistics，the Royal Statistical Society；Elected member of the International Statistical Institute。

主辦：研究生院
承辦：kaiyun開(kāi)云官方網(wǎng)站