Modeling Expected Returns with R

The main goal of the project is to reproduce selected key findings from the empirical asset pricing literature and related investment practices. Among our background references there are Ilmanen (2011) and Engle et al. (2016), both focused on central issues in the subject. The project pertains to the Quantitative Analysis of Active Portfolio Management and the three broad areas under which implementations fall are: (i) approaches to dynamic asset weighting, (ii) return factors and their risk premia and (iii) time-varying expected returns. An integral part of the project consists in collecting financial data of interest, including Profs. Fama-French's academic library, replication data sets underlying research produced at companies such as AQR or Ibbotson Associates (Morningstar) and data from providers like Bloomberg, CRSP, and MSCI Barra. Real-world quantitative data analyses follow both to reproduce target papers and show functionality. Our work is bundled into an open source R package, released on GitHub under the AGPL. We believe that providing a modeling and validation toolbox will add value for the users of the field by allowing them to freely produce and reproduce research.