Representing a matrix as a JSON object is a task that appears in many modern data science contexts, in particular when one wants to exchange matrix data online. There is no universally agreed way to achieve this task and various options are available depending on the matrix type and the programming tools and environment one has available.

Matrices are in general not “native” structures in computing environments but are handled with speficic packages (modules, extensions or libraries). In this post (first of a small series) we will discuss a number of options and the associated tradeoffs for common current setups in python, R, julia and C++.

In subsequent posts we will explore additional issues and/or go in more depth in specific topics. The cumulative results will be collected in the matrix2json repository

In the latest Open Risk White Paper (OpenRiskWP11_021221) “Sustainable Portfolio Management: Attribution and Allocation of Greenhouse Gas Emissions", we discuss the different required information layers, approaches to emissions accounting, attribution and forward-looking limit frameworks implementing carbon budget constraints. The white paper serves also as a first installment of the mathematical documentation of the Equinox Software Platform (code repository here).

Dear Academy users,

we are very pleased to launch a new platform to help knit the Open Risk community closer together:

It is called the Open Risk Commons and it is a Discourse based forum, a popular platform that used by thousands open source communities around the world.

The aim (and hope) is to bring together people who are interested / involved in Academy courses but also the Open Risk Manual, Open Source Risk Models and Tools, or any of our more recent initiatives around Open Risk Data and Sustainable Finance.

NB: The Commons is a separate platform, so your Academy credentials will not work (a new self-registration account is required).

Hope to see you all there and please share also with people you know who might be interested to join and invent the open future of risk management!

Philippos,

Open Risk

Object-oriented programming and techniques (OOP) such as using classes and inheritance are common in many application programming environments but alas don’t “travel well” outside computer memory. The potentially intricate relationships of objects (both the data they hold and the meaning and possible uses of the data) are not easy to transfer (except of-course by full replication of code and data). Hence when considering data science tasks and objectives that involving exchange of data, the transition from object hierarchies that live inside memory, to data structures that can be exchanged with another computer is not straightforward. In this new short course we explore how some popular programming languages, data formats, database API’s and web frameworks handle hierarchical classes.

Incidentally, we have also wrapped up the GSOC 2021 mentoring project and a blog post summarizes developments.

We have started exploring options for making the Open Risk Manual available offline.  In a new blog post we discuss the plans and what is currently implemented, while a repository will be collecting such resources. Currently there is a draft collection of entries about data privacy.

If you have any specific ideas or recommendations about offline strategies please let us know!

In this blog post we discuss a number of financial terms whose precise meaning is frequently intentionally or unintentionally obscured. As a result those terms may, like a Rorschach Blot, mean different things to different people. Unlike this famous psychological test, ambiguity in weighty financial matters can have adverse consequences.