Testing A New Stock-Picking Model WASHINGTON - Anyone with a computer and data feeds can come up with an investment model. We normally don't pay much attention to such methodologies unless they've been evaluated over a long period of time and with real money at stake. But we came across a relatively new idea, known as SPF, that seems worthy of being put to the test. The inventor of this model isn't a newcomer to the investment business. Joseph S. Kalinowski, the director of research for New York brokerage Puglisi & Co., has six years of experience, five of which were spent analyzing databases at IBES International, an aggregator of analyst estimate information. Kalinowski has been with Puglisi since May. Kalinowski's SPF is an acronym for "sentiment," "persistence" and "fundamentals." In each of these categories, he ranks stocks according to a variety of metrics. Taken together, the rankings produce an SPF profile signaling whether to buy, sell or hold.
With the "sentiment" factor, Kalinowksi tries to predict how market psychology will affect a particular stock. In addition to watching price and volume trends, Kalinowski draws on his IBES experience to rank stocks according to one-month changes in analysts' earnings estimates for the current and next fiscal year, as well as the proportion of raised estimates to lowered estimates. The goal: to take advantage of the price "drift" that often occurs as analysts--who often flock together--revise estimates either up or down. The "persistence" part of SPF assesses quality of earnings, namely the degree to which a company's net results are backed up by cash flow and sales. Looking at quarterly financial history, Kalinowski measures the difference between net income and cash flow from operations. For long picks, the lower that number, the better. He also keeps an eye on how closely revenue growth correlates to growth in return on equity (net income divided by book value). Under "fundamentals," Kalinowski's formula generates an implied price for each stock. To do this, he flips the estimated price-to-earnings ratio to get the so-called earnings yield: the 12-month forward earnings estimate divided by the current stock price. He then compares the earnings yield to its historical average (up to ten years), considering also how closely earnings yield and stock price have correlated throughout that time frame. For example, suppose a stock's earnings yield falls 15% shy of its historical average. If there's 68% correlation between movements in earnings yield and stock price, Kalinowski is 68% sure the stock should be 15% higher. Keep in mind that all of Kalinowski's factors are interrelated. For example, his conclusions on implied price for a particular stock are tempered by earnings quality and market psychology towards that stock. We asked Kalinowski to apply the SPF model to our Platinum 400 list. The results--seven stocks with "favorable" profiles and six "avoids"--are in the tables below. One caveat: An "avoid" reading from Kalinowski's SPF model does not necessarily mean a company is awful; it simply suggests that the stock is ahead of where it should be trading.
'); //--> Subscriptions >
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Delivered By |
Tested By |
Market Data By |
Market Data By |
Market Data By |
American History |
Luxury Cars |