TMCnet News
Newfound Research Brings Out New White Paper on Robustness in Portfolio OptimizationApr 10, 2013 (Close-Up Media via COMTEX) -- Newfound Research, a Boston-based financial technology and quantitative investment product development firm, has released a new research paper exploring portfolio optimization, a highly debated asset allocation methodology among both retail and institutional investors. According to a release, the white paper, entitled Allocating Under Uncertainty: Simple Heuristics & Complex Models, is authored by Corey Hoffstein, the firm's co-founder and Chief Investment Officer. The research discusses and compares complex optimization methodologies to portfolios constructed from simple heuristics, such as an equal-weight methodology. The primary difference in the methodologies that the research focuses on is their usage of expected excess return, volatility, and correlation forecasts. Hoffstein argues that when a complex allocation methodology relies on non-stable input parameters, a simple heuristic-based method frequently outperforms. "Optimization methodologies may be mathematically proven to construct an optimal portfolio under several assumptions, but those assumptions are unrealistic in the real world," Hoffstein said. "Forecasts, which optimizations heavily rely upon, are inherently uncertain. For example, long-term equity forecasts of 8 percent are a best guess, but realized long-term averages may end up much lower or higher. Standard optimization methodologies cannot model this uncertainty. Not only can this uncertainty lead to dramatically different portfolio constitutions, but also can translate to significantly sub-optimal portfolios for future market environments." The analysis shows that when allocation methodologies ignore unstable forecasts, they demonstrate greater robustness in achieving their target goals on a walk-forward basis. "Added complexity frequently means reduced robustness," Hoffstein says. "Naive allocation techniques are actually an inherent hedge against input and model uncertainty." "This is probably the exact sort of research you would not expect a quantitative asset management firm to release," points out Tom Rosedale, co-founder and CEO of Newfound. "Releasing research contradicting the value-add of complex quantitative methods may seem unusual for us, but it is in line with our team's philosophy of quantitative integrity." More information: www.thinknewfound.com ((Comments on this story may be sent to [email protected])) [ Back To TMCnet.com's Homepage ] |