主题：From Statistical Learning to Financial Big Data Analytics

主讲人：统计学院 陈坤 教授

主持人：中国金融研究中心 潘志远 副教授

时间：2021年5月11日（周二）10:30-11:30

地点：格致楼618会议室

主讲人简介


陈坤，西南财经大学统计学院教授、博士生导师，光华英才学者，研究方向为时间序列分析、空间统计、函数型数据分析和金融统计。荣获光华优秀硕士学位论文指导教师。曾多次赴日本早稻田大学、日本北海道大学、台湾中研院、香港中文大学、浙江大学和其他国内外多所高校访问。主持并参与了多项国家自然科学基金项目、教育部人文社科项目;在《Insurance: Mathematics and Economics》、《Journal of Time Series Analysis》、《Electronic Journal of Statistics》、 《Canadian Journal of Statistics》、《Journal of Statistical Planning and Inference》等国际重要期刊发表论文；是《Statistica Sinica》、《Bernoulli》、《Journal of Time Series Analysis》、《Journal of Forecasting》等多个期刊的匿名审稿人。

内容提要

After the tumultuous period marked by the 2007-2008 Financial Crisis, the financial industry has entered a new era. Quantitative analysis, including statistical models and methods, and algorithms for their development and implementation, are of increasing importance. The onset of this era is “big data revolution”, where various modern statistical learning methods can be applied. In this talk, I will give a brief introduction on sparsity, LASSO and friends. Then two examples are shown to illustrate the application of these regularization methods to financial big data analytics. Finally, using a moderate training real data, these methods are shown effective in improving the performance of traditional financial models, algorithms and estimation methods.