This is a summary of links featured on Quantocracy on Wednesday, 06/15/2022. To see our most recent links, visit the Quant Mashup. Read on readers!
-
Fed Days: Pre vs Post-Announcement Action During Downtrends [Quantifiable Edges]In a blog post a few years ago I showed that the Fed Day edge has basically played out before the announcement even takes place. Returns after the announcement have been somewhat random. In last nights subscriber letter I decided to take a similar look, but only examining instances during long-term downtrends. Below is a look at how the SPY has performed from 2pm to 4pm on Fed Days where SPY
-
Optimization problems with non-continuous restrictions [Quant Dare]In the financial field, managers usually take advantage of the great development in machine learning techniques to improve their models and get the best performance of their portfolios. These techniques may be clustering, neural networks, or even a more traditional one as optimization algorithms. In general, the existing algorithms are suitable in most of the problems that managers have to face
-
Relative Sentiment and Machine Learning for Tactical Asset Allocation [Alpha Architect]By the middle of 2019, we had been running an ensemble of relative sentiment(1) indicators in live asset management for several years. One of the components of that ensemble was a strategy that looked at Sentix sentiment indices. For those unfamiliar with Sentix (a German company), every week it polls institutions and individuals separately about their current and future outlooks on various
-
Factors Investing in Cryptocurrency [Alpha Architect]Cryptocurrency investing is a widely debated topic and one can find plenty of debates on Twitter discussing the fed, fiat currencies, and inflation. Regardless of where you fall on the crypto spectrum, we try and focus on research-centric takes on various investment themes whenever possible. The authors of this study research the cross-section of cryptocurrency returns and ask the following: Are