统计与数据科学系系列学术报告之三百八十六-三百八十七

统计与数据科学系系列学术报告之三百八十六

时间：2023年03月03日（周五）15:00-16:00

主持人：复旦大学管理学院统计与数据科学系朱仲义教授

地点：李达三楼105室

报告人：林金官教授

江苏省政府统计与大数据研究院院长

南京审计大学统计科学大数据研究院院长

题目：Electricity consumption forecasting by a new neural network model: panel semiparametric quantile regression neural network (PSQRNN)

摘要：Addressing the forecasting issues is one of the core objectives of developing and restructuring of electric power industry in China. However, there are not enough efforts that have been made to develop an accurate electricity consumption forecasting procedure. In this paper, a panel semiparametric quantile regression neural network (PSQRNN) is developed by combining an artificial neural network and semiparametric quantile regression for panel data. By embedding penalized quantile regression with least absolute shrinkage and selection operator (LASSO), ridge regression and backpropagation, PSQRNN keeps the flexibility of nonparametric models and the interpretability of parametric models simultaneously. The prediction accuracy is evaluated based on China’s electricity consumption data set, and the results indicate that PSQRNN performs better compared with three benchmark methods including BP neural network (BP), Support Vector Machine (SVM) and Quantile Regression Neural Network (QRNN).

个人简介：林金官，男，博士，教授、博士生导师。江苏省政府统计与大数据研究院院长。现主要从事非线性统计、计量经济、金融统计与风险度量、统计诊断、面板数据分析和统计应用等方面的研究工作。2000年以来, 在国内外核心期刊上发表论文一百余篇，其中SCI和SSCI收录论文八十余篇。目前担任2018-2022年度教育部统计学类教学指导委员会委员、全国工业统计教学研究会副会长、《系统科学与数学》《数理统计与管理》《统计与决策》等杂志编委，其中《数理统计与管理》副主编。主持省部级以上课题18项，其中国家自然科学基金和国家社会科学基金5项。已培养博士生15人，培养硕士生数十人，另与8位博士后进行合作研究。

统计与数据科学系系列学术报告之三百八十七

时间：2023年03月03日（周五）16:00-17:00

主持人：复旦大学管理学院统计与数据科学系朱仲义教授

地点：李达三楼105室

报告人：王兆军教授

统计与数据科学学院执行院长

题目：Activation discovery with FDR control: Application to fMRI data

摘要：Data arriving in “streams” from a large number of sources is ubiquitous, a portion of which usually incurs structural changes during the time-course of data acquisition. For example, in fMRI analysis, some brain regions become active associated with task-related stimuli or even in resting-states. Such a region corresponds to an activated data stream. We are aiming to measure the uncertainty of discovering data streams in activation via the tool of the false discovery rate (FDR). Borrowing ideas from recent developments of the FDR control methodologies, we propose a simple yet effective method to achieve this purpose meanwhile taking unknown asynchronous change patterns and spatial dependence into consideration.Its validity on controlling the FDR is justified by asymptotic analysis. Numerical experiments indicate that the proposed method is both accurate and powerful. It is also applied in a real fMRI data analysis.

个人简介：王兆军，统计与数据科学学院执行院长/教授，教育部**学者特聘教授，国务院学位委员会统计学科评议组成员，中国工业与应用数学学会副理事长，中国工业统计教学研究会副会长，天津工业与应用数学学会理事长。曾任国家统计专家咨询委员会委员、中国现场统计研究会副理事长、天津市现场统计研究会理事长，曾获国务院政府特贴、全国百篇优博指导教师、教育部自然科学二等奖及天津市自然科学一等奖。