Examples of work
See below for:
* Certifications.
* Example reports.
* Modelling techniques.
See below for:
* Certifications.
* Example reports.
* Modelling techniques.
While a Master's student, I worked in the research department at Statistics Canada on LifePaths. I examined how the simulation model's immigrants compared to the experiences of real immigrants using the Census and Labour Force Survey data. The goal was to determine how well the model performed on these new Canadians.
See the report here.
During my Master's I came up with a twist on the traditional shift-share method. Rather than look at changes the level of employment across industries and regions, we looked at the change in occupations across regions. We focused on the regions of Alberta during a resource boom period. Our work resulted in a paper.
Brox, J., E. Carvalho, and J. MacKay. 2010. “Regional Employment Changes in a Booming Resource Economy: A Modified Shift-Share Analogue Regression of Changes in Employment Patterns within the Economic Regions of Alberta.” Canadian Journal of Regional Science 33(2): 25–44.
See my research page for more recent examples.
This is a regional economic analysis conducted after I graduated with my Master's in Economics, but before I started my PhD. My client was the York Regional Municipality in Ontario, Canada. The approach used is shift-share -- a popular approach to measuring regional economic growth. In this report, I measure changes in employment levels across different industries and regions of Ontario.
Here's the report.
Jupyter notebooks are a popular way to experiment with data and present results to business stakeholders. These two examples show how I used modern machine learning techniques for predictive analytics. In these Coursera examples, I use logistic regression, xgboost and random forest models to model two different example problems:
1) Gaining insight into employee turnover and predicting workers most at risk of leaving the company.
2) Determining which customers are more likely to stop using the Waze driving application.
This shows the economic relationships between different industries in New Zealand and China. The Power BI visualization allows for dynamic exploration of the graph.
I often work with networks. I thought it would be fun to create a dynamic visualization of the economic relationships between industries in New Zealand and China. The visualization uses OECD data from the international input output tables to show those ties.
You can see the dynamic visualization and read more about my process here.
This page gives a high level overview of a recent project that involved using a topic model to analyze large amounts of social media data.
See the overview page here.
