Research

Research Interests

Business Analytics, Model Transparency, Model Diagnostics, Discrete Data, Network Inference, Statistical Inference/Machine Learning in Insurance/Information System.

To fulfill business needs and regulatory requirements, my research focuses on developing statistical/machine learning methods to address the multifaceted challenges associated with model transparency . The goal is to understand the inner workings of complex models, thereby promoting more transparent, trustworthy, and interpretable data-driven decision-making models . I primarily work on problems that involve discrete data (e.g., binary, rating or count data) which amplify statistical challenges and call for new developments.

Published Papers

Liu, D., Zhu, X., Greenwell B., & Lin, Z. (2022), “A new goodness-of-fit measure for probit models: surrogate R2”, accepted by the British Journal of Mathematical and Statistical Psychology. https://doi.org/10.1111/bmsp.12289

Invited Revision Papers

Liu, D., Lin, Z., & Zhang, H. “A unified framework for residual diagnostics in generalized linear models and beyond”, invited revision at the Journal of the American Statistical Association.

Working Papers

Zhu, X., Lin, Z., & Liu, D., “Surr rsq: an R package for evaluating goodness of fit using surrogate R2”, manuscript for submission to the R Journal.

Lin, Z., Liu, D. & Li, J., “Tweedie modeling of insurance premium: breaking the box using diagnostics tools.”, in preparation for Management Science.

Lin, Z., Liu, D. & Samuel, B.,“Joint modeling of multivariate discrete outcomes? An exploratory framework and its application for the design of information system”, in progress.

Presentations

"Surrogate R^2: a new goodness-of-fit measure and an R package for categorical data analysis", Invited session, New England Statistics Symposium (NESS), Boston, MA., 06/2023

"Unfolding Tweedie model for insurance pricing: a diagnostic tool leading to actionable insights", Contributed poster, The Eighth Bayesian, Fiducial and Frequentist conference (BFF8), Cincinnati, OH., 05/2023

"Model diagnostics of discrete data regression: a unifying framework using functional residuals", student contributed poster, The Joint Statistical Meetings (JSM), Washington D.C., 08/2022

"Model diagnostics of discrete data regression: a unifying framework using functional residuals", refereed extended abstract, Symposium on Data Science and Statistics (SDSS), Pittsburgh, PA., 06/2022

"Model diagnostics of discrete data regression: a unifying framework using functional residuals", student award presentation, New England Statistics Symposium (NESS), Mansfield, CT., 05/2022

"Analyzing conflicting information via multi-dimensional textual network analysis framework", INFORMS Annual Meeting, Virtual., 10/2020

Teaching Experience

Instructor (In-person & Online)

Undergraduate Course

BANA 4085 Spreadsheet Analytics (2021 Spring)

Graduate Course

BANA 6043 Statistical Computing (2021 Fall)

BANA 7025 Data Wrangling (2022 Fall)

BANA 7046 Data Mining I (2022 Spring, 2023 Spring)

Teaching Assistant

BANA 2081 Business Analytics I

BANA 2082 Business Analytics II

BANA 4085 Spreadsheet Analytics

BANA 4143 Data Management for Analytics

BANA 4137 Descriptive Analytics and Data Visualization

BANA 6043 Statistical Computing

BANA 7052 Applied Linear Regression

BANA 7046 Data Mining I

BANA 7047 Data Mining II

Second Reader for Capstone Essays