I am an interdisciplinary quantitative scientist and economist based in Washington DC. I find interest in all things data, in all fields of application and decision-making. A model agnostic, I tend to use whatever tool I have, to solve a problem or answer a question. In my work I have borrowed tools/methods/techniques from traditional statistical, economics, econometrics and machine learning fields, to social science and Goegraphical Information Systems (GIS). I have worked in 4 different continents and my 20+ technical writings have appeared in international journals.

Mahbubul Alam, PhD
Forest economist turned data scientist; trained in environmental-economic modeling
- Washington DC
- Github
- Google Scholar