gingado: a machine learning library focused on economics and finance

Date: Thursday 26 May 2022, 5.00PM - 6.00PM
Location: Online
Online - joining instructions will be sent to those registered
Special Interests Group Meeting
Book now


Share this event

gingado is an open-source machine learning library designed to facilitate the use of machine learning models in economics and finance, with the goal to broaden the accessibility of these models to a broader range of practitioners. By offering a standardised API on top of the well-known scikit-learn library (and extensible to include other machine learning frameworks and techniques), gingado simplifies the user experience for its target use cases. The library has three main features. First, gingado automatically provides practitioners with reasonable benchmark models once the model is specified and the data is loaded. This enables users to assess candidate model specifications more efficiently.  Second, gingado includes methods for data augmentation, using SDMX and APIs methods to complement the dataset at use with the relevant publicly available data at the jurisdiction and/or time level from statistics sources such as the BIS, IMF and others. The data augmentation aims to contribute to improve performance of the machine learning model. Third, gingado automatically documents the machine learning models resulting from its use, which can be useful for model audits and forensics. Contributions from the community in the form of Python code or feature requests to gingado are welcome.
 
Speaker: Douglas K. G. Araujo

Douglas Araujo works in the Bank for International Settlements (BIS) in Switzerland, at the Secretariat of the Basel Committee on Banking Supervision, where he oversees global discussions on data governance, bank and supervisory uses of machine learning, among other themes. Before joining the BIS in 2018, Doug worked at the Central Bank of Brazil, where among other projects he was responsible for the creation of the Brazilian housing price index based on housing appraisal data. Doug was a Fellow at the BIS's Financial Stability Institute in 2014. From 2015 to 2018, he was also a member of the International Monetary Fund technical assistance missions to enhance countries’ macroprudential frameworks, including by helping authorities leverage regulatory data to calculate statistics such as housing price indices and firm income time series. Until 2011, Doug worked in the private sector in Brazilian financial markets. His current research focuses on bank and fintech lending behaviour during the Covid-19 crisis and the use of machine learning in economics.
 
 
Organising group: RSS Finance & Economics Special Interest Group 

Organiser: Dr Jia Shao
 
Book now