Business and Finance
About the Focus Area
Data science is at the center of informed business decision-making.
The global business landscape is increasingly data-driven. Valuable data is generated in every facet of business, and data science techniques help companies find pragmatic solutions to inform strategy, execution, and organizational leadership in finance, management, and marketing.
Data scientists recognize patterns and form insights, contribute to risk management, build models, create viable products, and present and communicate financial data to effectively support company executives in sales, product development, planning, and customer acquisition. Causal techniques, including correlation and multiple regression analysis, are used to prepare business forecasts and long-range plans. Financial analytics are used to audit and optimize financial strategies based on both company and industry data.
This nexus of data science and business analytics and Columbia’s position in the global business hub and ecosystem that is New York City places DSI in a unique position to leverage data science to improve business processes.
DSI researchers study revenue maximization problems and combine labor market data with data science methods to identify factors and environments that shape gender and racial inequality.
Teams of M.S. in data science students have developed and evaluated novel spatio-temporal trajectories clustering methods used for traffic planning, inventory optimization, and understanding movements; used public data, anomaly detection, and natural language processing to create a dashboard and pip-installable package to help executives learn the full news history of a potential client; and predicted the Moody’s rating changes.
Collaboratory courses also prepare Columbia Business School students to succeed in an increasingly data-intensive world with a curriculum focused on programming, databases, and data analytics and the latest techniques for gathering, managing, and interpreting data.
- This research team combines new sources of labor market data, which include online resumes and employee reviews, with data science methods to identify factors and environments that shape gender and racial inequality in the high-skilled labor market. The team charts long-term career trajectories of a large number of American workers, examines gender and racial variations, and constructs measures of company environment that pertain to gender and racial equity.
- Kriste Krstovski, Data Science Institute, Business
- Yao Lu, Sociology
- This course exposes real estate development graduate students to technology and programming foundations, and offers a set of electives that explore the use of these technologies and analytical tools in the real estate development domain. The courses are designed to proceed in tandem with detailed instruction in the related domain content, initially with foundational courses followed by lab courses around specific applications or methodologies. A Proptech capstone project is also offered in the final semester of the degree program.
- Patrice Derrington, Architecture, Planning and Preservation
Hardeep Johar, Industrial Engineering and Operations Research
- This series of new courses prepares Columbia Business School students to succeed in an increasingly data-intensive world with a curriculum focused on computer programming and the latest techniques for gathering, managing, and interpreting data. The cross-disciplinary curriculum, called Technology, Coding and Analytics for Columbia Business School, provides foundational courses in programming, databases, and data analytics, as well as industry-specific electives (digital advertising, sports analytics, et al.).
- Hardeep Johar, Industrial Engineering and Operations Research
- Costis Maglaras, Business
- Mentor: Tapan Shah, GE Research
- Team Members: Tabitha Karuna Sugumar, Fatima Koli, Jacqueline Araya, Kun Tao
With increasing use of asset tracking, large databases of spatio-temporal trajectories (STT) are used for traffic planning, inventory optimization, and understanding movements. This team developed and evaluated novel STT clustering methods with promising results. The key novelty arises from different similarity measures (DTW, Fréchet Distance, Edit Distance, LCSS) and representations (e.g. autoencoders) for STT.
The “Know Your Client” requirement sets a broad mandate that financial institutions must execute adequate background checks on potential clients. This team used public data, anomaly detection, and natural language processing to create a dashboard and pip-installable package to help executives learn the full news history of a potential client.
- Mentor: Naftali Cohen, JPMorgan Chase & Co.
- Team Members: Jesse Patrick Cahill, Thomas Causero, James Anthony DeAntonis, Ryan Owen McNally
- Are external factors able to affect long-term trends in sales? Answering this question is crucial for hind-sighting and determining future outlook of fashion stores. To explore the impact of events on sales, major event features were engineered to predict sales at the Orlando FOA store from April 2018 to September 2019. The team recommended to scale up the analysis to other stores to re-examine the effect of events, or switch the focus to other external factors like news to direct marketing strategies.
- Mentors: Rohit Cherian, Ralph Lauren; Ami Pascale, Ralph Lauren; Sining Chen, Data Science Institute
- Team Members: Qi Feng, Yawen Han, Jiaying Jin, Ruizhi Zhang
The announcements of Moody’s credit rating change have impact on the financial market. Predicting them in advance enables more informed investment decisions. The team was interested in predicting the next Moody’s rating change to generate alpha. They recommended to get more data (i.e. Moody’s ratings for more companies) and perform time series classification with a deep model.
- Mentors: Eunice Hameyie, Vanguard; Smaranda Muresan, Data Science Institute
- Team Members: Yunwen Cai, Aleksandra Hosa, Haoran Li, Vincent Pan
Kira Goldner (Ph.D. in computer science and engineering, University of Washington) does research on algorithmic mechanism design: designing algorithms that guarantee that the designer’s objectives are achieved, even when the data they run on is produced by individuals acting in their own self-interest. Her work includes studying revenue maximization problems, and understanding more behavioral and informational complex models.