In this article, a review of several methods for Factor Investing and Asset Allocation is carried out, with the intention of clarifying the differences and similarities between these two concepts that are currently the subject of debate in the Financial
Industry, particularly in Portfolio Management and Hedge Fund Management.
It is a fact that Asset Allocation and Factor Investing, despite being crucial activities to improve performance of investment portfolios and hedge funds, have been neglected by practitioners, whether investors or investment advisors.
In the academy, current literature almost forgot the specialists who remain in the line of fire. As Markowitz, founder of Modern Portfolio Theory, states in a book by Kinlaw et al.: “Our field has an abundance of articles by academics trying to persuade other academics as to how practitioners should advise clients.”
The reviews of methods for both Asset Allocation and Factor Investing done in this work have the purpose of disseminating these methods among practitioners.
Above all, it is intended to clarify that it is not about applying one or the other, but about knowing and applying both, if what is sought is to achieve better investments performance.
Asset Allocation is that blend of stocks and bonds that investors hold over the long term, while Factor Investing is a subfield of a large discipline that encompasses asset allocation, quantitative trading and wealth management. Its premise is that differences in the returns of firms can be explained by the characteristics of these firms.
Regarding Asset Allocation, the selection of assets using Higher Statistical Moments (Skewness and Kurtosis) taken from financial time series distributions is reviewed, the calculation of Cumulants by means of the Cornish-Fisher Expansion is also reviewed. As long time known, financial time series do not behave following the Normal Distribution, so consideration about this fact is needed in statistical computations.
Regarding Factor Investing, some Machine Learning tools, like Principal Components Analysis and Support Vector Machines are reviewed.