EconText Lab
About Us
Economics investigates questions ranging from how we choose a jar of jam in a store to global consequences of climate change. But the field overwhelmingly uses structured data representable in spreadsheets. As we increasingly consume written content online and offline, this form of data shapes our everyday choices, moves markets, and influences democracies. The field of natural language processing develops quantitative measures from unstructured text.
At EconText, we unlock new possibilities by combining the two fields. We bring the rigor of causal inference and statistics to bear upon textual data to answer important questions in innovative ways.
Projects
Skills of Machine Teachers
Machine learning (ML) and AI are all the rage in the age of ever-increasing data and compute power. But beyond these buzz words, do we know what skills employers truly demand? We decided to hear what the horses themselves say. Using texts of online job postings, we study the evolution of skills demanded by firms seeking AI and ML workers.
Demonetization Disruption
In November 2016, India's government unexpectedly declared 86% of currency bills in circulation no longer legal tender. A cash-dependent economy suddenly didn't have cash! How did firms navigate this disruption? We use firm financial data and annual reports to examine this question. Learn more here.
Diversity & Firm COVID-19 Response
The pandemic has left no business untouched. Some went bankrupt, some are struggling, and some have created opportunities. Did firms' diversity influence this fate? We use firms' narratives about COVID-19 challenges and actions to investigate how diversity at the top impacts the speed and quality of firms' crisis response.
Our team
We are a team of one computer scientist, two economists, and several students across majors and class levels.
Meet the Team
Join Us!
We are accepting applications from students interested in learning natural language processing, statistics, economic analysis, and coding skills through hands-on projects.