Carl H. Lindner College of BusinessCarl H. Lindner College of BusinessUniversity of Cincinnati

Carl H. Lindner College of Business

Yichen Qin, PhD

Assistant Professor
Professional Summary
Yichen Qin
Yichen Qin is an assistant professor in the Department of Operations, Business Analytics, and Information Systems in the Lindner College of Business at the University of Cincinnati. He earned his Ph.D. degree in Applied Mathematics and Statistics from the Johns Hopkins University in 2013. For more information, please go to https://sites.google.com/view/yichenqin/
Contact Information
E-mail:
Office:
529 Carl H. Lindner Hall
Phone:
513-556-7025
Fax:
513-556-5499
Teaching Interest
  • BANA2081 Business Analytics
  • BANA7050 Forecasting and Time Series Methods
  • BANA7038 Data Analysis Methods
Research Interest
  • https://sites.google.com/view/yichenqin/
  • Computational Statistics
  • Robust Statistics
  • Mixture Models
  • Network Analysis
History

Institution:
University of Cincinnati
Title:
Assistant Professor


Awards | Honors

Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2018


Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2017


Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2016


Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2015


Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2015


Organization:
University of Cincinnati, College of Business
Name:
Dean's List of Teaching Excellence
Year Received:
2013


Education

Institution:
Johns Hopkins University
Major:
Applied Mathematics and Statistics
Completed:
2013
Degree:
Ph D


Institution:
Johns Hopkins University
Major:
Financial Mathematics
Completed:
2012
Degree:
MS


Published Contributions

Yichen Qin, C. Priebe,  (2017). Robust Hypothesis Testing via Lq-Likelihood. Statistica Sinica, 1793-1813.


N. Zhou, W. Cheng, Yichen Qin, Z. Yin,  (2015). Evolution of High Frequency Systematic Trading: A Performance-Driven Gradient Boosting Model. Quantitative Finance, 1387-1403.


Y. Li, C. Yu, Yichen Qin, L. Wang, J. Chen, D. Yi, B.-C. Shia, S. Ma,  (2015). Regularized Receiver Operating Characteristic Based Logistic Regression for Grouped Variable Selection with Composite Criterion. Journal of Statistical Computation and Simulation, 2582-2595.


X. Zeng, S. Ma, Yichen Qin, Y. Li,  (2015). Variable Selection in Semiparametric Models for the Strong Hierarchical Longitudinal Data. Statistics and Its Interface, 355-365.


J. Li, D. Yi, Yichen Qin, Y. Shen, Y. Li,  (2014). Feature Selection for Support Vector Machine in the Study of Financial Early Warning System. Quality and Reliability Engineering International, 867-877.


Y. Li, Yichen Qin, Y. Xie, F. Tian,  (2013). Grouped Penalization Estimation of Osteoporosis Data in Traditional Chinese Medicine. Journal of Applied Statistics, 699-711.


Yichen Qin, C. Priebe,  (2013). Maximum Lq-Likelihood Estimation via the Expectation Maximization Algorithm: A Robust Estimation of Mixture Models. Journal of the American Statistical Association, 914-928.


D. Nehren, D. Fellah, J. Ruiz-Mata, Yichen Qin,  (2012). Dynamic Density Estimation of Market Microstructure Variables via Auxiliary Particle Filtering. Journal of Trading, 55-64.


Y. Li, D. Yi, H. Zhang, Yichen Qin,  (2012). Syndrome Evaluation in Traditional Chinese Medicine Using Second-Order Latent Variable Model. Statistics in Medicine, 672-680.



Accepted Contributions

A. Athreya, D. Fishkind, K. Levin, V. Lyzinski, Y. Park, Yichen Qin, D. Sussman, M. Tang, J. Vogelstein, C. Priebe,  (Accepted). Statistical Inference on Random Dot Product Graphs: A Survey. Journal of Machine Learning Research.




Research in progress

Title:
A User Similarity Network Based Approach for Distributed Recommendation Systems

Description:
Recommendation has now become an indispensable part of people’s everyday online experiences including viewing stream videos, listening to music, shopping online, and so on. However, serious issues on customer data privacy, recommendation security, and difficulty in generating cross-site recommendations arise with the centralized server-based architecture currently employed by online service and product providers. We propose a distributed recommender system based on the construction and maintenance of a user similarity network in which each user maintains only a small number of close neighbors for peer to peer (P2P) neighbor discovery and recommendation generation. Empirical results using the Netflix dataset show that our proposed system achieves comparable recommendation quality as the centralized recommendation with significantly fewer similarity calculations to identify user neighbor sets for recommendation generation. Various settings are tested on small-scale datasets to verify the practical designs of the distributed recommender system we proposed. A large-scale experiment also shows that our P2P recommender system has the potential to obtain high recommendation accuracy with very limited number of similar neighbors.

Status:
On-Going

Research Type:
Scholarly


Title:
Cluster-Based Predicitve Models in Online Education

Description:
As online education becomes a popular learning approach, the large amount of data generated by learning activities can be smartly utilized for the evaluation and assessment purposes. Traditional analytical tools and techniques are being adopted by the online education industry to improve services in critical areas such as student retention, grades, and graduation. To evaluate an online learning environmnent for students at the University of Phoenix, cluster analysis and regression analysis techniques were implemented to develop a cluster-specific predictive model and a simple direct regression model for student service management. In the cluster-specific predictive model, finite mixture models are used to classify students based on their learning attributes. Then, based on the learners segmentation framework, predictive models were developed to predict the target scores for given new learner attributes. Numerical results show that the cluster-specific model has a better performance for the model fitting. The advantages and limitations for each method are discussed and recommendations are provided for management to drive academic excellence.

Status:
Writing Results

Research Type:
Scholarly


Presentations

Organization:
ASA
Location:
Baltimore, MD
Year:
2017


Organization:
ICSA
Location:
Chicago, IL
Year:
2017


Organization:
ERCIM Working Group on Computational and Methodological Statistics
Location:
Seville, Spain
Year:
2016


Location:
Chicago, IL
Year:
2016


Organization:
University of Osijek
Location:
Osijek, Croatia
Year:
2016


Organization:
University of Zagreb
Location:
Zagreb, Croatia
Year:
2016


Location:
Geneva, Switzerland
Year:
2016


Organization:
Auburn University
Location:
Auburn, AL
Year:
2015


Organization:
University of Georgia
Location:
Athens, GA
Year:
2014