Factor Tilting Portfolios: Which is Superior Tilting Methodology?

Summary

Factor Tilting refers to a method of exposing a portion of a portfolio to a factor investing strategy. The quantitative factor calculation utilizes all the observable data such as stock prices, financial data, and forecasts for the company. This research performs and compares different factor tilting through several methodologies. How much a portfolio is exposed to a factor and how it will have effect is different according to different tilting methodology.

Prior to applying the factor tilting methodology, the robustness of well-known factors was checked. We continued the research using momentum factor, since performance trend was more pronounced as we divide stocks into more quantiles. For detailed evaluation of various tilting methodologies based on the momentum factor, we compare different strategies by examining excess returns, factor exposure, and turnover of each. All investigated tilting methodologies show positive IR values when studying its excess returns for the entire period, proving their advantage over pure passive investment. Among those, the signal tilting (ST) method was the most dominant with an IR of 0.1394, and the Capitalization Weight (CW) method shows the lowest at 0.0584. When looking at the factor exposure, all tilting methodologies showed positive and significant active factor exposure. Both signal weighting (SW) and signal tilting (ST) have the largest in terms of active exposure, while active risk constrained optimization (ARCO) and capitalization weight (CW) appears to have relatively low exposure. With turnover of a portfolio, capitalization weight (CW) was the lowest, and the signal tilting (ST) method showed the highest turnover.

We expect the result and the analysis of each method to help investors understand the characteristics and performance of each tilting method and help implement appropriate tilting methodologies. As mentioned above, various tilting methodologies can be used for various factors, and when multiple factors are used, a more diverse portfolio can be formulated. The analysis also shows that tilting methods provide better risk-adjusted return compared to the benchmark, so that higher performance can be pursued with the same level of risk. Recently, many global asset managers are releasing smart beta ETFs tailored to new themes and management philosophy of each manager. The trend suggests thatit may be necessary to pay more attention to the tilting strategy.