QuantStart News - August 2020

Find out what QuantStart has been working on in August 2020.

A couple of months ago we started a new set of posts designed to keep the QuantStart community aware of what the QuantStart team had been up to in previous month. In last month's post we discussed what we had been working on in July 2020.

Articles and Tutorials

In August we once again reviewed our Content Survey for 2020 and noted that while Machine Learning took the top spot in terms of votes there was a very healthy part of the community that were interested in Mathematical Finance.

Mathematical Finance is traditionally taught (at MFE/graduate level) via rigourous probability theory that leads into stochastic calculus. We thought it would be instructive to begin an article series that concentrates on the main themes of this rigourous approach, beginning with a look at a mathematical concept known as a sigma algebra:

We also continued the Deep Learning series of articles by explaining how to train the perceptron model that was outlined in last month's article on the same topic, using the Scikit-Learn and TensorFlow libraries in Python:

Behind the scenes we were also working hard on developing documentation for the QSTrader project. Over the coming months we will be adding more tutorials to help you get started more easily with backesting in Python.

Quantcademy

Along with new article content QuantStart has been answering many interesting questions from the Quantcademy membership forum community.

Topics discussed in August include installing MySQL on a modern Mac environment for securities master databases, how to utilise machine learning/deep learning to forecast equity prices at intraday frequencies along with interview preparation for targeting early stage quant researcher roles in the UK.

What's Next?

We will be attempting to craft some more detailed QSTrader tutorials in the current month in order to move beyond the elementary strategies we have currently presented. This will help us move beyond the statically-allocated portfolios presented into those which have more of a tactical flavour.