The aim of the course is to:

• familiarize the students with  real life problems in quantitative asset management,
• enable them to read, evaluate and reproduce research publication from major investment banks and asset manager,
• develop research proposals, implement and test them.

The course will involve steady empirical work and data analysis. The R software (base + relevant financial libraries) will be used extensively. An introduction and detailed explanations to the software will be given as the course will move from simpler to more advanced topics.

The course will evolve around two major topics:

1. Asset Allocation. The problem of allocating wealth among different assets in order to obtain an optimal mix of risk and expected return will be analyzed. The Markowitz Model will be presented and critically analyzed. This model in fact presents a series of drawbacks which hinder its practical use. We will consider several refinements and their implementation: benchmark, enhanced benchmark, 130/30 strategies, Black Litterman approach. In this context the concepts of strategic and tactical asset allocation will be introduced.
2. Forecasting Asset Returns. Active Asset Management is essentially about forecasting asset returns and developing "views" on assets. This problem will be analyzed through a series of case studies. Practitioners and academic papers on the topic will be presented for several asset classes (Equity, Credit, Interest Rates, Fx)

Weekly homework (75%) + Final Project and Presentation (25%)

Mixture of research publications from major investment houses (JPMorgan, Nomura, Citi, Societe' General...) and sound(er) academic work:

"La gestion d'actifs quantitative" Thierry Roncalli
http://www.thierry-roncalli.com/
(The book is in French however most of the research papers on his web page are in English)

"Risk and Asset Allocation" Attilio Meucci
http://www.symmys.com/

Campbell Harvey
http://www.duke.edu/~charvey/
http://www.duke.edu/~charvey/Classes/ba453/syl453.htm

Diethelm Wuertz et al.
https://www.rmetrics.org/

knowledge of statistics/econometrics and programming, interest in empirical work, willingness to participate actively.