Skip to content
View lambert-tan's full-sized avatar

Block or report lambert-tan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lambert-tan/README.md

Hi, I'm Lambert 👋

I like turning messy data into answers people can actually use.

I'm a Master of Management in Analytics candidate at Queen's University with a background in Statistics and client-facing banking experience at RBC and TD.

I moved into analytics because I enjoy the part between “we have data” and “so what should we do?” My projects usually start with a practical question, then work through the data, modelling, validation, and interpretation needed to answer it.

📍 Toronto · 📊 Business & Data Analytics · 🐶 Usually working with a Chihuahua nearby (Her name is Amiu btw :P)


What I'm working on

I’m especially interested in problems involving customer behaviour, pricing, risk, and business decision-making. I work mainly in Python, with experience across statistical modelling, machine learning, simulation, optimization, and BI.

🏙️ Toronto Airbnb Pricing Analytics

15,332 listings · Pricing · Statistical modelling · Streamlit

What actually drives Airbnb prices in Toronto?

I built a reproducible analysis around a log-price OLS model and found that property format and bathroom setup explain much larger price differences than smaller operational features. The final model explains about 61.8% of log-price variation on the test sample.

View the analysis · Try the live pricing app

🛒 Negative Review Prediction in E-Commerce

95,824 orders · Classification · Customer experience

How early can a business identify an order that is likely to end in a negative review?

I compared models at different points in the order lifecycle. The later CatBoost model performed better (ROC-AUC 0.768), while the earlier model gave the business more time to intervene. The interesting part was not simply which model scored higher, but when a prediction becomes useful enough to act on.

View the project

❄️ Toronto Warming Centre Optimization

Resource allocation · Monte Carlo simulation · Optimization

How should limited staffing and bed capacity be allocated when winter demand is uncertain?

I combined optimization with a 5,000-iteration Monte Carlo simulation to stress-test capacity across seven warming centres. The analysis showed how the operational bottleneck could shift from staffing to physical bed capacity.

View the project


Tools I use

Python · pandas · NumPy · scikit-learn · Statsmodels · LightGBM · CatBoost
SQL · R · Excel · Tableau · Power BI · Streamlit · Git

Methods: regression · classification · clustering · model validation · Monte Carlo simulation · optimization


A little more about me

Before analytics, I worked in client-facing banking at RBC and TD. That experience still shapes how I approach data: a technically correct result is only useful if someone can understand what it means and make a better decision with it.

Outside the notebook, I’m usually exploring Toronto, travelling, or spending time with my Chihuahua.

Currently

🎓 Master of Management in Analytics — Smith School of Business, Queen's University
🔎 Exploring opportunities in business analytics, data analytics, customer analytics, and risk analytics
📍 Toronto, Canada

Pinned Loading

  1. lambert-tan lambert-tan Public

    Config files for my GitHub profile.