Open Database

The Linkable Open Data Environment (LODE) is an exploratory initiative that aims at enhancing the use and harmonization of open micro data primarily from municipal, provincial and federal sources.

The results are a collection of datasets released under a single open data license (Open Government Licence - Canada), as well as open source tools used to process the data, and collaborations in an open space.

Built upon the LODE, I have compiled the provincial into countrywide data, and added FSA (first 3 digits of postal code) column using geospatial join, I have also added indicator to distinguish between condos and other type of structures.

The complete open address dataset is approximately 1.8G, so download when you are connected to WIFI ;)

Download

Machine Learning in Insurance

​• Geo_LLM: Address parsing using classification LLM for NER (Named Entity Recognition) and address correction using sequence LLM.

Software Development and Automation

Utility Management Tool Provides you with the following solution:
​• ResQ one-stop Automation Solution
​• Task Scheduler
​• Excel Batch Processing

Read More

Baby Activity Tracker

​• Record daily baby activities

Read More

LA FIRE Monitoring

The ongoing LA fire has captured everyone’s attention and hearts. I have been closely monitoring its growth, containment efforts, and the potential damage it could cause. As an actuary, I strive to ensure that our work helps people during such crises by advocating for sufficient reserves to support recovery. It’s about more than numbers—it's about enabling communities to rebuild and return to their homes with resilience and hope.

LA FIRE Monitoring
LA FIRE Impact Estimate As of Jan 14, 2025 (Number of Properties at Risk by Zip Code)
LA FIRE Impact Estimate As of Jan 14, 2025 (Number of Properties at Risk by Zip Code with Buffer)

Catastrophe Modeling

This catastrophe model consists of four key modules: Hazard, Inventory, Vulnerability, and Loss. The Hazard Module is a statistical model that reflects flood susceptibility across regions using a flood susceptibility map, Canada's historical flood map, and seasonality adjustments to account for temporal variations in flood risk. The Inventory Module represents the exposed assets, incorporating their location, value, and structural characteristics. The Vulnerability Module quantifies the relationship between flood intensity and asset damage using damage functions. Finally, the Loss Module translates asset damage into financial loss by incorporating policy conditions, reinsurance structures.

Sample Loss Summary
simulated_flood_events
simulated_flood_events_zoomin

Data Playground

I am thrilled to provide high-fidelity synthetic claims data produced by my Claim Simulator which is an advanced actuarial tool designed to provide actuaries with realistic, structured datasets for reserving, pricing, and risk analysis. Unlike traditional models that rely solely on historical claims, this simulator produces synthetic claim trajectories that reflect real-world development patterns while allowing for extensive scenario testing. At its core, the simulator features a self-developed distribution fitter, ensuring that frequency, severity, and development assumptions align with actuarial conventions. It incorporates copulas and multivariate correlation structures, allowing for realistic dependency modeling across claims, coverage types, and accident periods. The synthetic claims generated by the simulator can be structured into loss development triangles, which can be aggregated to quarterly or annual levels and converted between incremental and cumulative formats. This tool empowers actuaries by enabling stress testing, model validation, and scenario exploration beyond the constraints of historical data. By simulating synthetic claim emergence and development, it facilitates data-driven decision-making in reserving, reinsurance structuring, and portfolio risk assessment.

Incremental Incurred Claims Development Data
Cumulative Annual Incurred Claims Triangle