A partitioned quasi-likelihood for distributed statistical inference

Guo，Guangbao1; Sun，Yue1; Jiang，Xuejun2

2020

10.1007/s00180-020-00974-4

COMPUTATIONAL STATISTICS   影响因子和分区

0943-4062

1613-9658

In the big data setting, working data sets are often distributed on multiple machines. However, classical statistical methods are often developed to solve the problems of single estimation or inference. We employ a novel parallel quasi-likelihood method in generalized linear models, to make the variances between different sub-estimators relatively similar. Estimates are obtained from projection subsets of data and later combined by suitably-chosen unknown weights. We also show the proposed method to produce better asymptotic efficiency than using the simple average. Furthermore, simulation examples show that the proposed method can significantly improve statistical inference.

Distributed data Distributed statistical inference Parallel computing Projection matrix Quasi-likelihood

SCI

英语

其他

Natural Science Foundation of Shandong[ZR2016AM09] ; Natural Science Foundation of Guangdong[2017A030313012]

Mathematics

Statistics & Probability

WOS:000562488000001

SPRINGER HEIDELBERG

MATHEMATICS

2-s2.0-85081727212

Scopus

1.Department of Statistics,Shandong University of Technology,Zibo,255000,China
2.Department of Mathematics,Southern University of Science and Technology,Shenzhen,518000,China

Guo，Guangbao,Sun，Yue,Jiang，Xuejun. A partitioned quasi-likelihood for distributed statistical inference[J]. COMPUTATIONAL STATISTICS,2020.

Guo，Guangbao,Sun，Yue,&Jiang，Xuejun.(2020).A partitioned quasi-likelihood for distributed statistical inference.COMPUTATIONAL STATISTICS.

Guo，Guangbao,et al."A partitioned quasi-likelihood for distributed statistical inference".COMPUTATIONAL STATISTICS (2020).