Projects – Wenxuan Zhu

Check out my projects below to see some of my previous work using public data.

Causal Analysis of Debt-to-Income Ratio on Loan Default Risk

Causal Inference

Using instrumental variables and multiple imputation to identify the causal effect of debt-to-income ratio on loan default probability. This study addresses endogeneity…

ClustVarSelect: Robust Feature Selection for Single-Cell RNA Sequencing Data

Feature Selection

An R package implementing power k-means clustering with Bregman divergences and sparse manifold decomposition (SMD) for effective feature selection in high-dimensional…

Linear Threshold Model and Simulation on Random Graphs

Network Science

Curious about how a single idea can ripple through a network, sparking trends and shaping behaviors? The Linear Threshold Model (LTM) explores how influence spreads through…

Exploring Bayesian Additive Regression Trees (BART)

Statistical Algorithm

An exploration of Bayesian Additive Regression Trees (BART), showcasing its strengths in predictive modeling and uncertainty quantification through theoretical insights and…

Spatial Analysis of Urban Heat Island in Twin Cities & Phoenixs

Correlated Data

Urban heat island effect (UHI) poses significant health risks and varies across American cities. This report uses Spatial Autoregressive (SAR) and Conditional Autoregressive…

Long-Term Effects of Speed Processing Training on Older Adults

Correlated Data

Longitudinal Data

Using robust Generalized Estimating Equations (GEE) model, we identified statistically significant improvements in speed processing among adults who received training.

Time Series Analysis of Macalester Main Campus Electricity Usage

Correlated Data

Time Series Data

A case study of Macalester College in St. Paul, MN uses time series analysis to model and forecast energy use using the campus’ main meter.

Exploring CIC-DUX4 Sarcoma Surfaceome by RNA-Sequencing

Pediatric Cancer

Bulk RNA-Seq analysis

High-Performance Computing (HPC)

During summer intern at UMN, I investigated genes encoding surfaceome in CIC-DUX4 sarcoma, uncovering vital patterns through data analysis and dynamic visualizations.

Expectation Maximization Algorithm and its Applications

Statistical Algorithm

An interactive R bookdown website on the EM Algorithm, showcasing comprehensive content, coding examples, and practical applications.

Matrix Completion and Recommendation Systems

Computational Linear Algebra

Recommender System

How does Netflix always know exactly which movies you like? Uncover recommendation system using Applied Singular Value Thresholding (SVT) and Cosine Similarity.

Self-Perception and Art? A Causal Inference

Causal Inference

An in-depth assessment of the potential causal relationship between Self-Perception of Creativity and Participation in Arts using data set from National Archive of Data on…