This is a summary of links featured on Quantocracy on Friday, 08/13/2021. To see our most recent links, visit the Quant Mashup. Read on readers!
-
Embeddings of Sectors and Industries using Graph Neural Networks [Gautier Marti]You can find the reproducible experiment in this Colab Notebook. In econometrics and financial research, categorical variables, and especially sectors and industries, are usually encoded as dummy variables (also called one-hot encoding in the machine learning community). You can find plenty of such examples in the SSRN literature, where authors are regressing the performance of their signal on
-
Exploring the rsims package for fast backtesting in R [Robot Wealth]rsims is a new package for fast, realistic (quasi event-driven) backtesting of trading strategies in R. Really?? Does the world really need another backtesting platform?? Its hard to argue with that sentiment. Zipline, QuantConnect, Quantstrat, Backtrader, Zorro there are certainly plenty of good options out there. But allow me to offer a justification for why we felt the need to build
-
Community Alpha of QuantConnect – Part 2: Social Trading Factor Strategies [Quantpedia]This blog post is the continuation of series about Quantconnects Alpha market strategies. This part is related to the factor strategies notoriously known from the majority of asset classes. Although the results are insightful, they are not straightforward, and further analysis could be made. Therefore, stay tuned for the next parts! Introduction We have already established that we can construct
-
Research Review | 13 August 2021 | Market and Asset Analytics [Capital Spectator]Decomposing Momentum: Eliminating its Crash Component Pascal Bsing (University of Muenster), et al. July 15, 2021 We propose a purely cross-sectional momentum strategy that avoids crash risk and does not depend on the state of the market. To do so, we simply split up the standard momentum return over months t-12 to t-2 at the highest stock price within this formation period. Both resulting