Mingxuan Yang

Data Scientist

iHeartMedia

Biography

I am currently the data scientist at iHeartMedia. My master’s degree in statistics was obtained from Duke University in May 2021. This website is a summary of my skillset and past experience. If you have any questions, please feel free to contact me.

Thanks for visiting my website!

Interests

Machine Learning
Data Analyst
Data Visualization
Causal Inference

Education

Master's in Statistical Science, 2021
Duke University
BS in Mathematics, 2019
Wuhan University
BS in Economics, 2019
Wuhan University

Skills

Python

100%

R

90%

SQL

100%

Statistics

100%

Looker

100%

BigQuery

90%

Projects

German NMURx Dashboard

A DataFest project to analyze the non-medical use of prescription drugs in Germany.

Berry Yield Prediction

Applied time series technology to predict the berry yield in the next 10 weeks.

COVID-19 Analysis

Constructed a dashboard about COVID-19, its economic influence and the potential dietary method to prevent it.

Airbnb Price Prediction

Applied machine learning models to predict the price of Airbnb listing places in Buenos Aires.

PhD Dashboard

Explored doctorate-granting institutions and doctorate recipients in the US using plotly, dash, Chart Studio and Streamlit.

Fraud Detection

Detected fraudulent agents in financial payment services based on binary classification models.

Spotify SQLite3 Schema

Established a SQLite3 schema to store Spotify dataset following database normalization 1NF, 2NF and 3NF.

Malaria Visualization

Visualization of Malaria to explore its spread and severity using Python.

Recommendation Systems

Constructed movie recommendation systems based on Netflix datasets.

SGHMC

Construction of a Python package to efficiently implement Stochastic Gradient Hamiltonian Monte Carlo algorithm.

Duke Textbook

Implemented a survey to estimate the expected expenditures on required textbooks at Duke University.

Paris Painting Price Prediction

Predicted the prices of paintings in the 18th century Paris using predictive models.

Penalized Linear Regression

Built a Shiny dashboard to visualize the Penalized Linear Regression using R.

News Dashboard

Created a Shiny dashboard that serves as a central news hub.

Thesis

Analysis on the health inequality of Chinese elderly

The research data comes from Chinese Longitudinal Healthy Longevity Survey (CLHLS). This paper established the ill-health index based …

The pricing of convertible catastrophe bond exampled by earthquake catastrophe

The securitization of catastrophe risk can effectively transfer catastrophe risk to the capital market. This paper proposes an …

Recent Posts

Insect Image Classification Using CNN

This blog is to understand the Python implementation of Convolutional Neural Network (CNN) through the insect classification example.

Nov 13, 2020 7 min read Python, CNN, Deep Learning, Classification

Insect Image Classification Using CNN

Data Extraction Using Star Wars API

API is the abbreviation for Application Programming Interface, which enables data transmission between multiple software intermediaries. This blog is to extract data through Star Wars API in the Python environment.

Oct 17, 2020 6 min read Python, API, Database, JSON

Data Extraction Using Star Wars API

Creating a SQLite3 Schema Using Spotify Dataset

This blog is to create a SQLite3 schema to store Spotify dataset following the requirement of database normalization 1NF, 2NF and 3NF. The normalization design will reduce data redundancy and prevent inconsistencies on insertion, update and deletion.

Oct 7, 2020 9 min read Database, SQL, SQLite3

Creating a SQLite3 Schema Using Spotify Dataset

The Magic of Weighting in Causal Effect Estimation

The balance of observed covariates plays an important role in caual inference, since it is the foundation of an extremely useful result called model-free validity for model-based statistics. Inverse Probability Weighting and Overlap Weighting can be quite beneficial for covariate balance.

Sep 30, 2020 8 min read Causal Inference, Covariate Balance, Overlap Weighting, Inverse Probability Weighting

The Magic of Weighting in Causal Effect Estimation

Malaria Visualization

Malaria is a fatal disease caused by infected mosquitoes. This blog is aimed to gain more comprehension about this epidemic through several visualizations.

Sep 23, 2020 6 min read Visualization, Malaria

Malaria Visualization