講座題目：A scalable nonparametric specification testing in massive data（大數(shù)據(jù)非參數(shù)模型識(shí)別）

主 講 人：南開大學(xué)王兆軍教授

時(shí) 間：2016年12月16日（周五）13:30-14:30

地 點(diǎn)：6號(hào)學(xué)院樓415教室

主 持 人：羅季副教授

主辦單位：數(shù)據(jù)科學(xué)學(xué)院

摘 要：

Lack-of-fit checking for parametric models is essential in reducing misspecification. However, for massive datasets which are increasingly prevalent, classical tests become prohibitively costly in computation and its feasibility is questionable even with modern parallel computing platforms. Building on the divide and conquer strategy, we propose a new nonparametric testing method, that is fast to compute and easy to implement with only one tuning parameter determined by a given time budget. Under mild conditions, we show that the proposed test statistic is asymptotically equivalent to that based on the whole data. Benefiting from using the sample-splitting idea for choosing the smoothing parameter, the proposed test is able to retain the type-I error rate pretty well with asymptotic distributions and achieves adaptive rate-optimal detection properties. Its advantage relative to existing methods is also demonstrated in numerical simulations and a data illustration.

主講人簡(jiǎn)介：

南開大學(xué)統(tǒng)計(jì)研究院教授，教育部長(zhǎng)江特聘教授，國(guó)務(wù)院學(xué)位委員會(huì)統(tǒng)計(jì)學(xué)科評(píng)議組成員，中國(guó)現(xiàn)場(chǎng)統(tǒng)計(jì)研究會(huì)副理事長(zhǎng)，中國(guó)統(tǒng)計(jì)學(xué)會(huì)常務(wù)理事，天津市現(xiàn)場(chǎng)統(tǒng)計(jì)研究院理事長(zhǎng)，天津市統(tǒng)計(jì)學(xué)副會(huì)長(zhǎng)。主要研究方向?yàn)榻y(tǒng)計(jì)過程控制(SPC)、非(半)參數(shù)回歸、降維、高維數(shù)據(jù)分析、變點(diǎn)。主持國(guó)家級(jí)課題6項(xiàng)，曾獲天津市自然科學(xué)一等獎(jiǎng)、全國(guó)百篇優(yōu)博指導(dǎo)教師。

歡迎全校師生踴躍參加。