PRZOOM - /newswire/ -
Washington, DC, United States, 2009/11/10 - Sentrana, Inc. recently addressed the North American statistician community with a new approach to Principle Component Regression (PCR), a modeling technique that provides distributors and retailers with information for making pricing decisions.
Sentrana Inc., the Washington, D.C.-based leader in revenue optimization technology, recently addressed the North American statistician community with a new approach to Principle Component Regression (PCR), a modeling technique that provides distributors and retailers with unprecedented information for making pricing and cross-selling decisions. This technique addresses the marketing puzzle of discovering the unique characteristics of every customer, and then marketing to these subtle needs rather than focusing on commonalities. By understanding the specific qualities that make one customer different than another, companies can fine-tune their deals and promotions to drive market-share and customers’ share-of-wallet growth.
Sentrana’s research on Bayesian PCR (BPCR) was scrutinized at the Joint Statistical Meetings (JSM) 2009 conference –– the largest gathering of statisticians held in North America. Provided by the scientific and academically elite, the scrutiny of this new research’s scientific credibility was a critical step in validating BPCR as an innovative improvement to PCR. Senior Quantitative Modeler at Sentrana, Liuxia Wang, Ph.D., presented the company’s findings in her talk on Bayesian Principal Component Regression with Data-Driven Components Selection, which was favorably received by the scientific community and is now leading the research on BPCR.
Rooted in quantitative marketing, Dr. Wang’s research into BPCR has far-reaching implications. Through the application of her technique to a problem in stock-price forecasting, she determined which financial metrics are most closely watched by the market during quarterly earnings announcements of a Fortune 100 food distribution company, thereby proving the effectiveness of this new technique. Sentrana’s research also found that BPCR not only provides a better fit to historical sales data, but also provides a more accurate description of the impact of price, promotion, and seasonal changes on demand for each product. Dr. Wang’s results showing the greater accuracy of this approach held up to questioning by statistical experts.
BPCR hinges on taking into account the basic fact that much of a customer’s purchasing behavior is unobserved by any single company (since customer’s spread their purchases out over many companies) and this creates uncertainty around what truly distinguishes one customer from another. Bayesian mathematics provides powerful machinery for grappling with this uncertainty, whilst Principal Components provide equally powerful machinery for pinpointing the differences that are truly important out of a haystack of millions of differences. The integration of these two mathematical instruments marks an advance in the field of marketing science.
“A large amount of research to date has been focused on finding the general patterns of what customers have in common and then setting prices, promotions, and cross-sell offers according to those commonalities,” Dr. Wang said. “However, insight into what customer’s don’t have in common can be much more lucrative and powerful, especially as customer tastes and product variety continue to proliferate explosively. Fortunately, new scientific methods are proving critical in overcoming this prevalent problem.”
Sentrana’s MarketMover™ technology platform is able to utilize a number of innovative models developed by Sentrana through on-going research. MarketMover™ enables distributors of a wide range of products to set prices for entire geographies and inventories in line with hourly changes in supply, demand and purchasing behavior. Sentrana has already begun integrating BPCR technology within MarketMover, and is stress-testing it within several high customer-volume companies with sales ranging from $300M to $1B. General availability of MarketMover with BPCR is slated for the fourth fiscal quarter of 2009.
Sentrana, Inc. (sentrana.com), based in Washington, DC, is a leader in providing revenue optimization solutions in highly competitive B2B & B2C markets. We provide end-to-end revenue optimization solutions by combining advanced quantitative methodologies with qualitative managerial inputs into our patented MarketMover™ platform. Our technology facilitates optimization of the marketing decisions at the individual transaction level that drive financial performance, allowing organizations to actively shape their demand environments. Our solutions enable our clients to make better pricing decisions, increase profitability, and unify disparate business units under a single revenue strategy.