编辑:时间:2021-11-30 13:02:42 浏览次数:
Recently, the paper “Pseudo-Bayesian Classified Mixed Model Prediction” by Dr. Ma Haiqiang and Professor Jiang Jiming from the School of Statistics was officially accepted by the Journal of the American Statistical Association (JASA), a top international statistical publication.
The article “Pseudo-Bayesian Classified Mixed Model Prediction” proposed a new classification mixed model prediction method based on Bayesian ideas. This method can not only use network information to correctly match group labels, but also has good statistical properties for large samples. This paper proves that under certain conditions, the proposed method has consistency and asymptotic optimality in both group label matching and mixed model prediction. Simulation and actual data analysis verify the advantages of the proposed method.
JASA, a professional statistical journal published by the American Statistical Association,
is currently one of the top academic journals with the highest paper quality and the widest coverage in the international statistics community. It has successively published a large number of important statistical innovations. The inclusion of the faculty's papers of our college has a strong incentive and exemplary effect on the academic research of the college. The School of Statistics has always attached great importance to the introduction and training of high-level talents, paying particular attention to the growth of young teachers, and striving to advocate a positive scientific research atmosphere. The academic atmosphere in the school is strong and the teachers are sincerely united. In terms of contribution to scientific research, the college promotes a high degree of integration of discipline construction and local service, co-construction and sharing of scientific research innovation platforms,
In terms of scientific research contribution, the college promotes the integration of discipline construction and local service, the joint construction and sharing of scientific research innovation platforms, scientific research collaboration and achievement transformation. The college focuses on serving the national poverty alleviation strategy, and has been involved in poverty exit verification projects. It has won unanimous praise from the Poverty Alleviation Office of the State Council and local governments, and has been reported by authoritative mainstream media many times; in terms of the high-end think tanks, we have built a key think tank platform, improved the level of advice giving and government affairs consulting, strengthened the exchange of think tank talents, and focused on poverty measurement, poverty alleviation policy evaluation and other related theories and methods. The team has achieved outstanding results. Relevant papers and topics provide references for local economic construction, and a series of special reports provide important basis for decision-making and have been affirmed by leaders.
About the Author:
Ma Haiqiang, Ph.D., lecturer, graduated from the School of Management of Fudan University in 2016. His main research direction is functional data and quantile regression. He has published 8 SCI academic papers in statistical journals at home and abroad; he has hosted one youth project and one regional project from Natural Science Foundation of China, and three provincial-level projects.
Jiang Jiming, a distinguished professor of Jiangxi University of Finance and Economics, was selected as a Yangtze River Scholar of the Ministry of Education in 2016. He is currently a fellow of the American Association for the Advancement of Science (AAAS), the American Statistical Association (ASA) and the International Statistical Association (IMS). His main research directions are mixed effects models, variable selection, and biostatistics. He has successively served as the deputy editor of famous journals such as The Annals of Statistics, JASA, and Statistica Sinica. He has published more than 200 statistical academic papers and many monographs in statistical journals at home and abroad.