Q-Rank Methodology

What is it made of?
How Do You Use It? An Analogy
What It Isn’t
Technical Jargon- how we set the weight of each item
How We Decide What to Follow

What is it made of?

The Q-Rank is a combination of nine fundamental and technical models for each stock. The scores range from 99 (highest) to 1 (lowest); i.e., a stock with a Q-Rank of 99 has combined qualities better than 99% of the stocks measured. It combines a Technical Sub-rank (TSR) and an Earnings Sub-rank (ESR) which offer further insight into the qualities influencing the total Q-Rank. We would buy stocks with a Q-Rank ranks above 70 (even better though if Q-Rank is above 90) and avoid those below 50.

The Technical Sub-Rank (TSR) combines four models rewarding price leadership, trading status, long-term trend, and sector attractiveness. The relative strength indicator compares the price performance of each stock over the past four quarters to that of all 2,500 stocks in our Q-Rank universe. A relative strength score of 99 indicates that the stock's price performance in the past four quarters was better than 99% of the quantitative universe. Although the relative strength score looks at leadership for the past four quarters, the most recent quarter gets increased importance. Return reversal reflects the tendency of stocks to get overbought or oversold. It is for short-term traders only. It is based upon a stock's price change over the preceding two weeks. Stocks with higher returns in the prior two weeks get low scores because they tend to "reverse" and under-perform over the subsequent two weeks. Return reversal scores greater than 90 suggest potential short-term out-performance. Stocks with scores below 10 have usually skyrocketed recently and are likely to consolidate their gains, falling back to their long-term trends. A price/200-day moving average ratio is used to identify a stock's trend, and sometimes whether investors have discounted its price too optimistically or pessimistically¹. It is positive above 120%, negative below 90%. The industry combo model measures the attractiveness of a company’s respective industry based on consensus Wall Street earnings revisions and surprise, price/relative strength, price/book ratio, price/cash-flow ratio, price/sales ratio, and dividend yield. We find the top 25% of these industries attractive and would tend to underweight the bottom 25%.

The Earnings Sub-rank (ESR) combines value and growth characteristics divided into four categories. It rewards or penalizes companies based on Wall Street consensus earnings revisions, earnings surprises, earnings consistency, earnings growth/acceleration, and reasonable valuations.

Earnings surprise is the difference between the most recent actual reported earnings and Wall Streets' consensus earnings estimates (standard SUI calculation). Earnings Revision measures the past month change in Wall Streets' consensus earnings estimates of the company's one-year earnings estimate. Earnings acceleration identifies companies with rising earnings growth over the previous four years that will likely generate rising earnings growth rates for the next two years according to consensus expectations. It rewards companies demonstrating above trendline rates of growth. Earnings consistency identifies companies that have positive earnings over the prior three years that demonstrate low earnings volatility. It is a ranking of two combined parameters: 1) The number of quarters in the last 12 in which the company's earnings were positive; 2) The volatility of the earnings stream during the period. Return on equity identifies companies with an ROE significantly different from the overall universe. EPS Growth is based on consensus Wall Street earnings estimates for the company's next fiscal year. P/E is the price/earnings ratio based on next year's consensus earnings estimate. Growth to P/E compares the average expected earnings growth rate over the next two years to an average P/E based on those earnings.

How Do You Use It? An Analogy (Back to Top)

The Q-Rank model is like a research assistant: It allows you to follow a larger universe of stocks. This assistant reviews 3,000 stocks every night and reflects an opinion in a consistent, objective, and easy-to-understand form. The assistant allows you to focus on more detailed, specific research. The assistant gives an independent second opinion on your own buy/sell candidates, regularly surveys your stock watch list, identifies new buy/sell candidates, screens for specific factors such as earnings revisions, and delivers ideas within sectors or industries. The Q-Rank is a starting point that identifies stocks to which you may then apply your own experience, insights, judgment, and knowledge.

Q-Rank can improve timing by delaying purchases or sales that may be just a little too early. Q-Rank stocks ranked above 90 tend to significantly outperform the market; those under 10 tend to significantly under-perform. We believe investors considering buying a stock with a Q-Rank under 50 should delay purchase unless the stock offers an extremely solid value, new product and/or turnaround story. Investors considering selling a stock with a Q-Rank over 90 should consider holding the stock until its rank declines. The Q-Rank is meant to assist pure fundamental research, not replace it. We always recommend sound fundamental judgment in addition to our quantitative approach.

Regardless of style, the overlay of a Q-Rank rank should improve performance. For example, a value investor can use the Q-Rank to stay in value stocks that have performed so well they may no longer represent good value but may just be hitting a strong momentum phase.

What It Isn’t (Back to Top)

Q-Rank is not a "paint-by-numbers" for managing money. The Q-Rank performance results illustrate the predictive power and risk of the model only. Used in isolation, the Q-Rank model could result in high turnover rates. Also, the Q-Rank ranks will often reflect concentrations in specific industry groups and economic sectors. Our experience shows the Q-Rank can help you by complementing—not replacing—your other stock research.

Technical Jargon- how we set the weight of each item (Back to Top)

Factor weights are computed based on relative return spreads over the last twelve months. We use the analytic hierarchy process (AHP) which is based on the concepts and techniques of hierarchical structuring, pair-wise comparisons, redundant judgments, and application of eigenvalues and eigenvector methodology to derive weights and consistency measures. The objective of pair-wise comparisons is to help the decision-maker evaluate the weights on a rational basis. The redundant pair-wise comparisons enable accurate capturing of the decision-maker’s subjective evaluations.

How We Decide What to Follow (Back to Top)

The Q-Rank includes equities traded on the NYSE, AMEX, and NASDAQ, but excludes those with (1) market capitalization below $200 million; (2) a current price less than $6 per share; and (3) insufficient financial information. These initial criteria are flexible. Our goal is to keep the Q-Rank stock universe to a manageable size—about 2,500 securities.

For more details, email a request to us.