Investment Research

Our Monthly Reading List

Our research is based on an extensive number of readings. About 33% of our working time is dedicated to studying and reading. Every month we screen many journals in search of valuable ideas to adapt and apply to our models. We focus our attention mainly on academic publications in the following journals:

  • Journal Financial Economics
  • Journal of Finance
  • Algorithmic Finance
  • Review of Financial Studies
  • Econometrica
  • Review of Finance
  • Journal of Financial Econometrics
  • Journal of Financial and Quantitative Analysis
  • Journal of Portfolio Management
  • Journal of Financial Markets
  • Journal of Forecasting
  • Journal of Finacial Data Science

After that, we constantly read books on investing and algorithmic trading written by practitioners. These authors provide an excellent integration of the academic literature. While this is not the place to provide a comprehensive list, we limit ourselves to the books read in the last six months:

  • “Algorithmic Trading & DMA”, 2009, by Barry Johnson
  •  “Algorithmic Trading Methods”, 2020, by Robert Kissell
  • “Machine Learning for Asset Managers”, 2019, by Lope de Prado
  • “Permutation and Randomization Tests for Trading System Development Algorithms in C++”, 2020, by Timothy Masters
  • “Statistically Sound Indicators for Financial Market Prediction”, 2019,  by Timothy Masters
  • “Machine Learning For Algorithmic Trading”,  second edition, 2020, by Stefan Jansen
  • “Technical Analysis for Algorithmic Pattern Recognition”, 2016, by Tsinaslanidis and Zapranis.
  • “Trading Evolved”, 2019, by Clenow
books-441866_1920-min

Understanding the Rationale

Before we proceed with the statistical analysis and the backtesting, we want to ensure that we understand why a given strategy should work from an economic standpoint. The reasons may be different and related to the market microstructure or behavioral dynamics. Anyway, we constantly want to make sure that there is a precise reason why a strategy should work. This is often ignored by institutional investors, who indulge in data mining often with disappointing results in the long term. Consequently, we try to avoid data mining as much as possible.

Only when our ideas on the economic rationale underpinning predictability are apparent are we confident in our trading System.

filter-1990470_1280-min

A Rigorous Statistical Approach

ruby-370130_1280-min
  • All our models must pass a dozen statistical tests before being considered successful.
  • This approach to avoiding false discoveries complies with the highest academic standards.
  • The complete list of statistical tests and procedures is detailed in the section dedicated to presenting our backtesting engine. 
  • The successful accomplishment of this stage is mandatory before proceeding to the paper trading and monitoring.
LEARN MORE