Investment Approach

While we would agree that it’s difficult to eke out an informational edge in today’s market in terms of what you know, we believe that how you use that knowledge can set you apart. Gerstein Fisher’s investment approach is about putting research and knowledge into action in ways that translate into real results.

As depicted at left, our investment strategy represents the convergence of several research streams:

Academic theory—The core tenets of our investment methodology are rooted in time-tested academic research. From findings on key long-term return drivers to explanations of enduring risk/reward relationships, we apply this knowledge daily to the practice of investment management.

Client knowledge—Equally important to investment success as proven research concepts and advanced statistical modeling techniques is a very real understanding of the individuals and institutions with whom we work. Only by truly knowing our clients can we make the other knowledge we possess relevant to their specific situations.

Investor behavior—While textbook descriptions of Modern Portfolio Theory seem cut and dried, in reality, investors are human beings who are subject to emotions. By understanding the common cognitive biases of investors, we can help our clients navigate investment decisions with the necessary objectivity to make sound, informed choices. In addition, we understand that markets, while reasonably efficient, are not always “rational.” As such, we’ve structured our investment process to quantify and capture proven investment factors that are at least partially behaviorally driven (for example, small cap, momentum and value).

Statistical methods—While we typically design portfolio structures to meet long-term objectives, the reality is that the path toward those objectives is rarely, if ever, “straight and narrow.” That’s why we use statistical techniques like Monte Carlo Simulation (MCS) to perform rigorous scenario analysis on portfolios before finalizing structure. We recognize that MCS itself is not an exercise in certainty, but rather in probability or possibility—and that it also entails tradeoffs: often portfolios with the highest confidence levels would require a client taking on too much risk or giving up too much by way of return potential. While we recognize their limitations, we believe that statistical techniques like MCS can play an important role in helping prepare our clients for a wide range of possible investment outcomes. Other examples of statistical methods and techniques that are part of our “toolkit” include regression analysis¹, multi-factor analysis², and return-and style-based analysis³.

¹ Regression analysis examines the statistical connection between a variable of interest and one or more factors used to explain its variation. For example, if the variable of interest is student test scores, regression could be used to show the connection to factors such as time spent studying or IQ.
² Multi-factor analysis of financial data is an application of regression analysis where statistically identified “factors” (such as the return characteristics of US small company stocks) are used to calculate the exposure of a given investment or strategy to these specific sources of risk and return.
³ Return based analysis is a statistical technique that estimates a fund’s style by determining the mix of passive benchmarks (e.g., the S&P 500) that best matched the actual returns of the fund over a specific time period. This analysis provides a historical perspective of how a fund has behaved, and how it might behave in the future. It is important to remember that for all of its strengths, returns-based style analysis is an estimate and it is not tied to the actual holdings of a mutual fund or sub account.