Optimization
Resampled Efficiency™ optimization treats investment information more realistically.
Risk-return estimates are, in essence, assumptions about the future. Traditional optimization processes treat these assumptions as 100% certain. New Frontier enhanced the traditional optimization process to account for this problem, earning four patents along the way. Our patented Resampled Efficiency utilizes a Monte Carlo technique to produce multiple sets of statistically-equivalent risk-return estimates based on the original estimates. Then, we take those estimates and use them to compute multiple efficient frontiers. This patented process explores the many ways assets in markets may perform given the original estimates and your confidence level in the estimates. The figure below shows various possible efficient frontiers developed from statistically equivalent risk-return estimates. Notice how different each alternative efficient frontier is, though they start with the same risk-return estimates. This demonstrates the variation possible when you are working with probabilities. Now suppose we have two investors who both want high risk portfolios, which would be to the far right on each of the efficient frontiers. Investors could land very different portfolios, even though portfolios started from statistically equivalent portfolios.
New Frontier develops its efficient frontier by averaging all of the frontiers developed from the statistically equivalent risk-return estimates. For example, the high risk portfolio of the Resampled Efficient Frontier is created by averaging the high risk point on all the efficient frontiers shown in the figure. Resampled Efficiency is an averaging process that distills all the alternative efficient frontiers into a new efficient frontier and set of optimized portfolios. These portfolios display enhanced return, effective diversification and managed risk. Comparing portfolio composition maps of a traditional efficient frontier and a Resampled Efficient Frontier with the same assets illustrates the advantages of RE. The traditional efficient frontier presents jagged transitions between risk-levels. The RE frontier includes more assets and avoids the large allocations to one or two assets that damage the diversification of the traditional portfolios. In particular, RE prevents the single asset portfolios that frequently appear at the low and high risk levels of the traditional efficient frontier.
Harry Markowitz, Nobel Laureate and the father of traditional optimization and modern portfolio theory, decided to test Resampled Efficiency in 2003. He, along with colleague Nilufer Usmen, pitted traditionally developed portfolios against Resampled Efficiency portfolios in a simulation test. To Markowitz' surprise, the Resampled Efficiency portfolios performed better in all 30 simulations even when the MV optimizer used better estimates for a starting point. The tests showed: first, that Resampled Efficiency outperforms traditional optimization techniques; second, that the Resample Efficient Frontier manages expectations by promising less though it delivers more; and third, that a better optimizer may often be more important than improving the risk-return estimates. This remarkable study strengthened perceptions of the power of Resampled Efficiency optimization to enhance investment portfolios.