A GSOC 2021 summer project collaboration between Open Risk and the Hydra Ecosystem Summer is underway and for the Google Summer of Code 2021 season Open Risk is happy to join forces with the Hydra Ecosystem. The project aims to guide students to build a hypermedia enabled REST service around standardized credit portfolio data. More specifically the project will build a REST service as backend for a hypothetical banking entity that collects and disseminates credit portfolio data conforming to an established public standard (the EBA NPL templates, see below).
Equinox is an open source platform that supports risk management and reporting of Project Finance. The platform integrates geospatial information with applicable regulatory and industry standards from EBA, PCAF and Equator Principles to provide a holistic view of the footprint of both individual projects and portfolios of project finance investments. Motivation Sustainability (understood in environmental, economic and social terms) is emerging as an undisputed constraint that will shape future human activity and more specifically how the financial system facilitates and empowers economic life.
Semantic Web Technologies integrate naturally with the worlds of open data science and open source machine learning, empowering better control and management of the risks and opportunities that come with increased digitization and model use The ongoing and accelerating digitisation of many aspects of social and economic life means the proliferation of data driven/data intermediated decisions and the reliance on quantitative models of various sorts (going under various hashtags such as machine learning, artificial intelligence, data science etc.
Celebrating Pi Day 2021 Pi Day is celebrated every year on March 14th. The reason of course is that the day is denoted in some calendars as (3/14), which evokes of 3.14, the first three digits of “π”. A thin excuse maybe but sufficient for the true believers to join along! The occasion represents an annual opportunity for mathematics and science enthusiasts to recite the infinite charms of Pi, including its irrationality, to talk to friends and family about math and its uses, and, when everything else fails, simply eat pie.
Course Content: This CrashCourse is an introduction to semantic data using Python. It covers the following topics: We learn to work with RDF graphs using rdflib We explore the owlready package and OWL ontologies We look into json-ld serialization of RDF/OWL data We try data validation using pySHACL We use throughout a realistic data set based on the Credit Ratings Ontology Who Is This Course For: The course is useful to:
The Non-Perfoming Loan Ontology The Non-Performing Loan Ontology is a framework that aims to represent and categorize knowledge about non-performing loans using semantic web information technologies. Codenamed NPLO, it codifies the relationship between the various components of a Non-Performing Loan portfolio dataset.(NB: Non-performing loans are bank loans that are 90 days or more past their repayment date or that are unlikely to be repaid, for example if the borrower is facing financial difficulties).
Constructing a Global Mobility Index (GMI): In previous posts (here, and here) we introduced new Open Risk Dashboard functionalities that integrate COVID-19 community mobility data (currently focusing on the datasets provided by Google). As a reminder, these reports chart over time human mobility trends collected from mobile geolocation data. The granularity is by geography and across different categories of places / activities such as retail and recreation areas, groceries and pharmacies, parks, transit stations, workplaces, and residential areas.
The Risk Function Ontology The Risk Function Ontology is a framework that aims to represent and categorize knowledge about risk management functions using semantic web information technologies. Codenamed RFO codifies the relationship between the various components of a risk management organization. Individuals, teams or even whole departments tasked with risk management exist in some shape or form in most organizations. The ontology allows the definition of risk management roles in more precise terms, which in turn can be used in a variety of contexts: towards better structured actual job descriptions, more accurate description of internal processes and easier inspection of alignement and consistency with risk taxonomies (See also live version and white paper (OpenRiskWP04_061415)
Using Sankey Diagrams: Sankey Diagrams are a type of flow diagram composed of interconnected arrows. The width of the arrows is proportional to the flow rate. Sankey diagrams are often used in physical sciences (physics, chemistry, biology) and engineering but also in economics. They can be used to represent the relative role and significance of various inputs and outputs in a given process. Sankey diagrams emphasize the major transfers within a system.
openNPL 0.2 release: The open source openNPL platform supports the management of standardized credit portfolio data for non-performing loans. In this respect it implements the detailed European Banking Authority NPL loan templates. It aims to be at the same time easy to integrate in human workflows (using a familiar web interface) and integrate into automated (computer driven) workflows. The latest (0.2) release exposes a REST API that offers machine oriented access using, what is by now, the most established mechanism for achieving flexible online data transfers.
Risk Management will not be the same going forward: too much is at stake! The summer is over in the Northern Hemisphere - and what an unusual summer has it been! Worldwide the implications and challenges of adjusting to a Covid-19 pandemic are still a major issue, affecting individuals, companies and governments. At Open Risk we have been tracking and will continue to interpret the impact of the pandemic via a number of projects:
openNPL now Available in Dockerized Form: Following up on the first release of openNPL the platform is now available to install using Docker. Running openNPL via docker is the installation option that simplifies the manual process (but a working docker installation is required!). Docker Hub You can pull the latest openNPL image from Docker Hub (This method is recommended if you do not want to mess with the source distribution).
Non-Performing Loans: The covid-19 crisis will certainly impact the concentration of Non-Performing Loans but given the special nature of this economic crisis compared (in particular) with the 2008 financial crisis it is unclear how precisely things will evolve. In a previous post and white paper (OpenRiskWP07_022616) we discussed the importance of advancing open and transparent methodologies for managing the risks associated with such credit portfolios. Effective management of NPL is also a top regulatory priority.
The community mobility reports and OpenCPM: In a previous post we introduced new OpenCPM functionality that integrates COVID-19 community mobility data (currently from Google). The reports chart movement trends over time by geography, across different categories of places such as retail and recreation, groceries and pharmacies, parks, transit stations, workplaces, and residential. While these reports are unlikely to persist as open data sources in the long term, the current availability (as of May 2020) enables providing within OpenCPM a mobility data dashboard that can help draw insights through visualization and statistical analysis.
The community mobility reports and OpenCPM: As the COVID-19 pandemic unfolded technology providers (most notably Google and Apple) made available to the public aggregated and anonymized data about human mobility in the crisis period (on the basis of smartphone location data). These Community Mobility Reports provide insights into how mobility patterns changed in response both to pandemic news and policies aimed at combating COVID-19. The reports chart movement trends over time by geography, across different categories of locations and activities, such as retail and recreation, groceries and pharmacies, parks, transit stations, workplaces, and residential.
Course Content: This course is a CrashProgram (short course) introducing the GeoJSON specification for the encoding of geospatial features. The course is at an introductory technical level. It requires some familiarity with data specifications such as JSON and a very basic knowledge of Python Who Is This Course For: The course is useful to: Any developer or data scientist that wants to work with geospatial features encoded in the geojson format How Does The Course Help: Mastering the course content provides background knowledge towards the following activities:
Course Content: This course is an introduction to the concept of credit contagion. It covers the following topics: Contagion Risk Overview and Definition Various Contagion Types and Modelling Challenges The Simple Contagion Model by Davis and Lo Supply Chains Contagion Sovereign Contagion Who Is This Course For: The course is useful to: Risk Analysts across the financial industry and beyond Risk Management students Quantitative Risk Managers developing or validating risk models How Does The Course Help: Mastering the course content provides background knowledge towards the following activities:
Making Open Risk Data easier: In an earlier blog post we discussed the promise of Open Risk Data and how the widespread availability of good information that is relevant for risk management can substantially help mitigate diverse risks. The list of Open Risk Data providers, particularly from public sector, keeps increasing and we are aiming to document all available datasets in the dedicated page of the Open Risk Manual.
Semantic Web Technologies: The Risk Model Ontology is a framework that aims to represent and categorize knowledge about risk models using semantic web information technologies. In principle any semantic technology can be the starting point for a risk model ontology. The Open Risk Manual adopts the W3C’s Web Ontology Language (OWL). OWL is a Semantic Web language designed to represent rich and complex knowledge about things, groups of things, and relations between things.
The motivation for federated credit risk models: Federated learning is a machine learning technique that is receiving increased attention in diverse data driven application domains that have data privacy concerns. The essence of the concept is to train algorithms across decentralized servers, each holding their own local data samples, hence without the need to exchange potentially sensitive information. The construction of a common model is achieved through the exchange of derived data (gradients, parameters, weights etc).