SOURCE: Common Sense Investment Management, LLC

Common Sense Investment Management, LLC

February 25, 2015 12:00 ET

Common Sense Investment Management Director of Research Develops New Method to Identify and Protect Insights Among Terabytes of Data

Findings Published in Journal of the Royal Statistical Society, One of the Top Journals in the Field of Statistics Since the 19th Century; Schneider's Work Helps Guide CSIM Investment Processes and Develop New Products, Reducing the Influence of Outliers and Identifying More Reliable Relationships in Large Data Sets; Research Sponsored by the National Science Foundation to Find Solutions That Prevent Leakage and Provide New Theoretical Guarantees

PORTLAND, OR--(Marketwired - February 25, 2015) - Common Sense Investment Management announced that its Director of Research, Dr. Matthew J. Schneider, has been published as the lead author in the prestigious Journal of the Royal Statistical Society, one of the top journals in field of statistics since the 19th century. His ground-breaking work, "A new method for protecting interrelated time series with Bayesian prior distributions and synthetic data", was co-authored with John Abowd, the Edmund Ezra Day Professor of Economics at Cornell University, and sponsored by the National Science Foundation (NSF). Bayesian techniques combine both quantitative data and qualitative judgments in a probabilistic way to limit future choices to help find reasonable connections amongst the tens of thousands of relationships between investments.

Dr. Schneider's research explores methods of identifying and protecting key insights in large data sets. He developed a new method to identify and protect insights among terabytes of data with millions of time series. The purpose of the NSF sponsorship was to foster collaboration between university-based research teams and major data-driven agencies to support the agencies' data privacy objectives. The research team found that the current theoretical models were inadequate at providing guarantees in data sets and preventing leakage.

"Identifying robust methods is just the first step in creating a sustainable data-driven process," noted Dr. Schneider. "Next, it is important to balance the tradeoff between the ability of users to generate persistent value-added insights and the tendency of users to follow the influence of outliers leading to highly unusual or unsustainable insights. My collaboration with major data-driven agencies was aimed at concealing these highly unusual insights, which is similar to my current work at CSIM. Investment teams want to avoid investing in unsustainable insights that are likely to disappear in the future."

Dr. Schneider also has papers in print at the International Journal of Forecasting, the Journal of Privacy and Confidentiality, and the International Conference of Data Mining

Dr. Schneider continues: "In order to protect large scale data sets, you first need to identify the persistent relationships that should be preserved, and the highly unusual insights that must be protected. Then, if your statistical methods are effective and scalable, the technology can be applied to varied domains. In support of the agencies' objectives, I created large 'fake' data sets for public release which preserved important persistent relationships while concealing confidential insights. Not only will the research support fund managers in tightening their investment processes, it will also support anyone who is concerned about protecting their data through data brokering, sharing, or public dissemination."

Dr. Schneider is a Visiting Scholar at the Samuel C. Johnson Graduate School of Management at Cornell University. Prior to joining CSIM, Dr. Schneider worked as an adjunct policy analyst for the RAND Corporation for five years, researching and advising on policy decisions for multi-billion dollar government programs. Prior to the RAND Corporation, Dr. Schneider served in the U.S. Navy as a Surface Warfare Officer on the USS Boxer. He holds a Ph.D. and M.S. in Statistics from Cornell University, a M.S. in Management & Public Policy from Carnegie Mellon University, and a B.S. in Quantitative Economics from United States Naval Academy in Annapolis.

For more information on Common Sense Investment Management, LLC, please visit: www.csimllc.com

About Common Sense Investment Management, LLC

Founded in 1991, Common Sense Investment Management, LLC ("CSIM") is an independent, privately-held investment manager located in Portland, Oregon. CSIM is a veteran alternative investments solutions provider with a 24 year track record of outperformance, portfolio differentiation and capital preservation. 

Contact Information