报告题目： Sparse Laplacian Synthetic Regression Model for Portfolio Allocation
报告人：Professor Arief Gusnanto (英国利兹大学）
邀请人：黄磊
讲座时间：2026年4月9日（星期四）10:50-11:30
讲座地点：犀浦校区三号教学楼 X30423
报告摘要：Current sparse solution to estimate the weights for portfolio allocation does not take into account that some stocks may be correlated. This is generally well known where e.g. tech stocks perform similarly in the recently. This makes the allocation challenging because the sparsity creates difficult interpretation. Current methods to address this challenge such as elastic net, fused lasso, or group lasso may have some limitations. In this talk, I will describe a proposed method to create a sparse solution where the dependencies of stocks are taken into account in the estimation. This is still an ongoing work with Dr. Lei Huang.
主讲人简介：Dr. Arief is an Associate Professor in the School of Mathematics, University of Leeds, UK. His research interest are in the development of statistical methods and inference in generalized linear and additive models with random effects, mixture models, and models to obtain sparse solution. He is the member of Royal Statistical Society and International Biometric Society.
