Factor investing is broadly described as allocating to systematic strategies that historically have delivered higher risk-adjusted returns than the market. Examples of factors used in practice are for instance momentum, carry, value, quality, low-risk, and flow. Although factor investing initially emerged within individual stock selection, nowadays the popularity of factor investing has spread to many other asset classes and investment instruments (see Asness et al., 2013).
With respect to equity factor investing on an aggregate level, models on country allocation have gained by far more attention than models on sector allocation.
Equity sectors form an attractive universe for active investment strategies. Companies within a sector share common operating characteristics and produce comparable products or services – take for example firms in finance, technology, ICT, energy, or utilities etc.
Across these sectors, we observe stark differences in both the degree (and the timing) to which they are sensitive to the business cycle and economic factors. The wide range in factor sensitivities, in turn, implies a large dispersion in the financial performance of sectors.
A factor investing strategy can benefit from differences in relative sector performance. Depending on the SIC-level (defining the sector granularity) the investment universe consists of a decent number of sectors. This increases the breadth of the active strategies. Moreover, the strategies can be implemented in a cross-sectional way, where the long/short set-up reduces the sensitivity to the overall equity market while allowing to maximally profit from relative performance differences.
In short, we believe that sector (or industry) allocation could be of added value next to country allocation. As choosing between Europe or the US is a totally different decision than choosing between IT or utilities, we could expect low correlations from both investment strategies’ alphas.
The paper by Doeswijk and Van Vliet (2011) will be the starting point of the research project. Your task is threefold: (1) extend their study with recent data and analyze these out-of-sample results, (2) enhance the current factors as defined in their paper, and (3) you are challenged to find new factors.
A specific direction we would like dive into for finding new factors will be real business cycle models. These models could be valuable for this research objective as different sectors tend to behave differently across the business cycle. See Lansing (2018) and Avdiev (2016) for some recent and interesting applications of such models.
 Asness, C.S., Moskowitz, T.J. and Pedersen, L.H., 2013, “Value and momentum everywhere”, The Journal of Finance, 68(3):929-985
 Doeswijk, R., & van Vliet, P. (2011). ”Global Tactical Sector Allocation: A Quantitative Approach”, Journal of Portfolio Management, 38(1):29:47
 Lansing, Kevin J., 2018. "Real Business Cycles, Animal Spirits, and Stock Market Valuation", Federal Reserve Bank of San Francisco, Working Paper Series 2018-8
 Avdjiev, S. (2016). “News Driven Business Cycles and data on asset prices in estimated DSGE models”, Review of Economic Dynamics, 20:181-197