How did you decide to enter the field of data science?
I started learning how to code for fun at the start of the pandemic; my primary interest was in data visualization and web app development. In June 2020, when my job in the nonprofit sector was eliminated, I had a lot of free time and decided to study data science because I’ve always loved statistics and storytelling. I followed my curiosity and quickly became fascinated with machine learning, then my obsession with language drove me to explore NLP in depth.
Since I am committed to social justice, I was naturally attracted to working with low-resource languages like Arabic. This led to an opportunity with the World Bank, where I was able to assist with social media research focused on the Middle East and North Africa. Working with economists sparked an interest in causality, and this was the reason I started to teach myself econometrics.
Later I was fortunate to be contracted to explore how NLP could be used for economic analysis, with a focus on sustainable development. I decided that I needed domain knowledge to better approach the problem; hence, I started to teach myself macroeconomics and development economics. Along the way I branched out to explore other areas that caught my interest, for instance linguistics, disinformation, cybersecurity, neuroscience, epidemiology, and AI ethics.
For the uninitiated, could you share a bit about what econometrics is, and why you find it interesting?
To me, econometrics is simply a set of tools that can be used to address causal questions. I was drawn to econometrics because I wanted to do more than make predictions using data-driven pattern recognition—I wanted to understand the “Why.” Causal inference is fascinating, and traditional econometrics usually provides the best methods to determine causal effects.
Moreover, machine learning is currently advancing econometrics in cool ways, and thanks to big data it is now possible to ask new types of predictive economics questions. Additionally, we now have the opportunity to apply a causal approach to data science, which I hope will improve the reliability and fairness of AI.
Understanding how to deal with pandemics is another reason to consider econometrics — epidemiology relies on the same underlying causal math as economics. Both disciplines require a very high degree of rigour owing to their direct impact on people’s lives. Good data is needed for ethical social policy, and causality is necessary to…
Continue reading: https://towardsdatascience.com/why-data-scientists-should-stay-open-minded-curious-and-non-judgemental-49784a883a78?source=rss—-7f60cf5620c9—4