If the process of generating earnings estimates was a science, we would be able to eliminate four columns from this table; the Number of Estimates, High Estimate, Low Estimate and Standard Deviation. But however skilled an analyst may be and however closely he/she communicates with management and studies data, one can never be 100% certain of any estimate and different analysts will arrive at different estimates. These four columns show you the range of opinion among analysts. As you review them, do so with the following principles in mind: a wider range of estimates means greater disagreement among analysts; greater disagreement reflects increased uncertainty; and Wall Street tends to dislike uncertainty.

On the one hand, you may choose to favor companies for which analyst estimates are close to one another. That would certainly put you on the "right" side of the principle that Wall Street dislikes uncertainty. But if such a company reports weaker-than-expected results, it will be taken by Wall Street as a harsher surprise and the stock might suffer more dramatically than would be the case if uncertainty (as indicated by a wide range of estimates) had been prevalent beforehand. There is no completely foolproof way to resolve this dilemma. Our advice: unless you are choosing to be a contrarian (in which case you might seek out wide estimate ranges), favor narrower estimate ranges unless other indicators point to significant stock overvaluation relative to industry peers. Proper portfolio diversification should help offset the impact of the periodic (and in a practical sense, unavoidable) earnings disappointments.

When studying the range of estimates, keep these points in mind:

• Standard deviation tells you about volatility. For example, the numbers 7, 8, and 9 average out to 8; so too do the numbers -7, 8, and 23. In a large sample, about two-thirds of all observations will fall one standard deviation above or below the average. So if you are looking at 100 numbers that have an average of 9 and a standard deviation of 6, about 65% of those numbers will fall between 3 and 15. If the standard deviation is 12, you'd have a much broader sample, with approximately 65% numbers falling between -3 and 21.
  
  
• Note the number of analysts. At this time the average number of analysts for the stocks in the Market Guide database is 5. But if we confine our inquiry to those stocks that are included in the S&P 500, the average number of analysts jumps to 15. A large number of analysts can add a measure of confidence to the estimates. But a smaller number of analysts can raise the possibility of a stock being a diamond in the rough; a relatively undiscovered company whose stock could perform especially well as more investors learn the story (assuming it's a good one).

Finally, the last column of the table offers two forward-looking versions of the stock's P/E multiple; one is calculated by dividing the current stock price by estimated earnings for the current fiscal year, and the other uses earnings for the next fiscal year. (This is in contrast to the standard P/E ratio in which we divide the current stock price by earnings for the latest twelve-month period.) You can compare these forward-looking P/E ratios to the consensus growth rate. Dividing the P/E by the growth rate will result in a PEG (Price/Earnings-to-Growth) ratio. Many interpret PEG ratios above 1.00 (P/E ratios that are above the growth rate) as signifying stock overvaluation. Be aware that under this yardstick, many (perhaps most) stocks nowadays are overvalued. So if you want to commit a significant portion of your investment funds to equities, you will probably find it difficult, if not impossible, to strictly use a 1.00 PEG threshold. But assuming you examine several stocks all of which have PEG ratios above 1.00, you can still use PEG to help you value stocks relative to one another (i.e. assume that a stock having a PEG ratio of 1.25 is being more reasonably value than an equity with for which the PEG is 1.80).