Columbia University Department of Statistics
2019 Minghui Yu Memorial Conference

About Minghui Memorial Conference

The 2019 Minghui Yu Memorial Conference, organized by doctoral students in the Statistics Department of Columbia University, will take place on Saturday, April 27th at the Faculty House. Minghui Yu was a doctoral student at the statistics department, who passed away in a tragic accident in the spring of 2008. Since then, doctoral students in the statistics department have been organizing a conference each year to honor his memory; this year will mark the 10th anniversary since the inaugural conference.

The conference will feature talks by doctoral students at the Statistics Department, ranging from those just beginning a research program to those who are about to defend dissertations. In addition to being an occasion to remember our friend and colleague, this event will be an opportunity to learn about exciting new research areas emerging from our department. We would like to thank the Department of Statistics for their continuous support.

About Minghui Yu

Minghui was born in Shandong, China in 1983. In 2002, he entered the Special Class for the Gifted Young at the University of Science and Technology of China (USTC), one of the most prestigious universities in China. Minghui possessed the rare quality of being not only smart, but also diligent, versatile, modest and easy-going. He was the type of friend who would stand by you no matter the situation. Minghui breezed through the challenging undergraduate program at USTC, ranking at the top of his class. Minghui was well liked by his fellows students having served as the class president from his sophomore year. Although under enormous academic pressure, he still found time to organize a series of student activities, such as hiking, art performances, and athletic contests for his fellow students.

After graduating summa cum laude in 2006 from USTC, Minghui entered the PhD program at the Physics Department of Columbia University. After one year, he transferred to the doctorate program in statistics. During his time at Columbia, Minghui served as the public relations head of the Columbia University's Chinese Students and Scholars Association (2007-2008), and was a member of the Columbia Chinese Basketball Association and the Columbia Graduate Student Consulting Club. His biography on the CUCSSA website mentioned his love of ''movies, photography and delicacies''. Minghui described himself in his blog as ''a boy who wants to combine art and science together''.

On April 4, 2008, after attending a student-organized conference, Minghui escorted his girlfriend home on the west side of campus. On his return, he was accosted as he was crossing 122nd and Broadway and in his attempt to flee, he was struck by an automobile on Broadway. Minghui was taken to St. Luke's Hospital where he passed away a short time later.

Keynote Speakers

We are excited to have two former graduates from the doctoral program in Statistics at Columbia University - Johannes Ruf and Tyler McCormick - as our keynote speakers this year. As students, both were part of the organizing committee for the inaugural Minghui conference, both received the Howard Levene Teaching Award from the department, and both held various service roles within the department.

Johannes is currently an Associate Professor in the Department of Mathematics at the London School of Economics, with his research focusing on stochastic analysis and its applications to mathematical finance. He began his education at the University of Ulm and Texas A&M University, later earning his PhD in Statistics under the supervision of Ioannis Karatzas, before becoming a Senior Research Fellow at the University of Oxford. He has also been a lecturer and a reader at University College London.

Tyler is currently an Associate Professor of Statistics and Sociology at the University of Washington, with some of his research interests being in the analysis of social networks and the estimation of demographic indicators with sparsely sampled data. He started his education as an undergraduate at Duke University, completing a masters degree at the University of Connecticut before obtaining his PhD in Statistics under the supervision of Tian Zheng, and has held various positions at the University of Washington since. He is currently the editor of the Journal of Computational Statistics and Graphics.

Program Schedule

For the full program schedule and abstracts, see the program book for details.

  • 8:30 - 8:55 Breakfast
    8:55 - 9:00 Opening remarks by Richard Davis
  • Morning Session I

    Chair: Cindy Rush

    9:00 - 9:20 Sihan Huang, Spectral Embedding of Optimal Weighted Adjacency Matrix (SEOWAM)
    9:20 - 9:40 Alessandro Grande, Overlapping Community Detection in Networks
    9:40-10:00 Andrew Davison, Asymptotic analysis for embeddings of exchangeable graphs
    10:00-10:20 Chengliang Tang, Large-scale Spatial Predictive Modeling with Applications to Ecological Remote Sensing Data
  • Morning Session II

    Chair: Tian Zheng

    10:40 - 11:00 Ian Kinsella, A penalized matrix decomposition for denoising and compression of functional imaging data
    11:00 - 11:20 Ding Zhou, A Spike+Gamma model for post-deconvolved calcium imaging traces
    11:20-11:40 Wenda Zhou, Towards Theoretically-Founded Learning-Based Denoising
  • Keynote Presentation I

    11:50 - 12:30 Johannes Ruf, Diversification, Volatility, and Surprising Alpha
  • Keynote Presentation II

    1:30 - 2:10 Tyler McCormick, Estimating spillovers using imprecisely measured networks
  • Afternoon Session I

    Chair: Sumit Mukherjee

    2:20 - 2:40 Nabarun Deb, Fluctuations in the mean-field Ising model
    2:40 - 3:00 Pratyay Datta, Optimal confidence bands under shape restriction in multi-dimensions
    3:00 - 3:20 Promit Ghosal, Multivariate quantiles and ranks using optimal transportation: Consistency and pointwise rate of convergence
  • Afternoon Session II

    Chair: Bodhi Sen

    3:40 - 4:00 Alejandra Quintos Lima, From Cox Processes to Fair Micro finance Loan Rates
    4:00 - 4:20 Bridget Ratcliffe, Using High Dimensional Chemical Abundance Data to Better Understand Milky Way Disk Formation and Evolution
    4:20 - 4:40 Zhi Wang, Latent feature extraction for process data
    4:40 - 5:00 Yuling Yao, Stacking for multimodal posterior distributions


If you would like to attend any of the talks, please email Andrew Davison.


The Faculty House is located at the intersection of 116th Street and Morningside Drive - please try and enter via the gate on Morningside Drive. The conference will be held in Garden Room 2 on the 1st floor, with lunch and dinner in the Presidential Ballroom on the 3rd floor.