THERE ARE A

lot of factors to weigh when it comes to buying or selling a stock. There's a stock's dividend yield, market capitalization, book-to-market ratio, momentum, analysts' recommendations, earnings forecasts and so on. The list is endless.

Wouldn't it be nice to know what the most important variables are to those who have devoted their careers to studying financial markets? And if Warren Buffett wasn't available, whose brain might you want to pick?

For Colby Wright, assistant professor of finance at Central Michigan University and James Doran, finance professor at Florida State University, it was other finance professors. After all, they're arguably the most educated and well-informed people when it comes to understanding the mysteries behind stock price movements.

So Wright and Doran set out to survey all the professors of finance in the U.S. and ask what's most important to them when investing their own money. The survey resulted in 642 usable responses. They published their results earlier this year in a paper titled "What Really Matters When Buying and Selling Stocks?"

The researchers' objective was not to explain why prices move in one direction or another, but to see whether "the models and theories developed to explain the risk-returns relationship of asset prices mean anything to those investors who actually trade," they write.

"I basically ask them: Out of all the stuff you've read, all the literature, what do you really believe?" says Wright. And whether you agree with why these academics buy or sell a stock, note that the paper says nothing about how successful the professors are with their investments just what they consider to be the key factors when deciding to buy or sell.

In other words, do these experts practice what they preach?

The answer turns out to be no. Out of 43 variables given, the most important were a company's price/earnings ratio and how close a stock is to its 52-week high to low. Considering the material most finance professors teach their students as a way of explaining stock price movements like the capital asset pricing model and discounted cash flows Wright calls the findings surprising.

SmartMoney.com: How are the results of your survey of finance academics important for regular investors?

Colby Wright: The real question we asked is: We've got all this academic literature out there that talks about what should you look at when you buy and sell stocks. What do the real experts look at when they buy and sell stocks? Hopefully, that will be useful to any investor. These professors are the most highly educated, highly sophisticated, they have high statistical training. They're familiar with the research that's been done.... Our ultimate objective was to ask what do finance professors really base their investing decisions on. There's a huge body of research in our field and there's so much conflict in the literature, it's hard to boil it down to what you should and shouldn't use.

We look at what might be helpful to help you beat the market. We emailed all the finance professors in the country...with a list of variables and asked how important are these things to you when you buy and sell stocks. There's been kind of a raging debate in academic circles is beta the only thing that matters? Then there's a whole camp that says you should look at P/E ratios, investor sentiment and dividend yields. We got some surprising results.

SM: What were the most surprising?

CW: The traditional valuation techniques and traditional asset pricing models when I say traditional I mean techniques that discount dividends, that use discounted cash flows, CAPM, arbitrage pricing theory, Fama and French, stuff we've taught to undergraduate and graduate classes that stuff ranks at the bottom of the list. That's basically the laundry list of what we've been teaching people.

We ended up focusing on professors who were trying to beat the market, and who trade stocks at least monthly. So these are not totally ivory tower academics. I call them active traders. When I focused on that group, the things at the top of the list, things they look at most were P/E ratio and market cap essentially, the size of stocks. After that, it's all related to momentum, which includes the stock's return over the past six months and over the past year, and the stock's 52-week high and 52-week low. So if you're doing a horse race of traditional asset pricing models vs. market anomalies, these guys trade on anomalies.

SM: Why are the momentum-related variables considered a market anomaly?

CW: If there is some piece of information or some strategy that generates returns not attributable to risk, this is considered an anomaly, since return is supposed to be driven by risk. There has been a disagreement in academia about momentum. Many people, myself included, believe that the trading rule suggested by momentum long winners, short losers does not have an adequate risk-based explanation. Therefore, the returns it generates, which have shown to be fairly robust, are considered anomalous. The fact that an investor could implement some sort of trading strategy based solely on looking at the returns to stocks over the past six months or year and earn a return greater than the risk he bears is almost unfathomable. Yet, the finding has withstood the test of time over the past 13 years.

SM: Wouldn't P/E ratio be considered a traditional valuation technique?

CW: You could argue that. We just ask how important these variables are to you when you buy or sell a stock.... But the literature that specifically led us to ask this question was a finding that stocks with low P/E ratios perform better than stocks with high P/E ratios. According to that you'd want to buy stocks with low P/E ratios. I didn't include that as a traditional valuation technique.

What's interesting about the use of P/E to value a stock or make some sort of investment decision is that use of a P/E ratio including use of the P/E valuation model really has little theoretical support, particularly if you accept markets as more or less efficient. True, P/E ratio is an indication of the cost of the stock [i.e. share price] relative to the value [i.e. the earnings], but efficient markets should price the value into the stock, so there should be no information in the P/E ratio. Yet, finance professors consistently report that it is the most important variable in their eyes when investing.

SM: Why was it surprising to find the momentum variables near the top of the list?

CW: The idea behind momentum...is basically buy winners, sell losers. Sell the stocks that have done poorly over the last six months and buy the stocks that have gone up over past year. Go long winners, go short losers. It's in direct opposition to contrarian strategy. Some huge hedge fund and mutual fund managers, maybe they don't make decisions based on momentum but they probably look at it before they hit the enter button. And these finance professors said it's among the top variables they consider when buying stock. Our question was: How important was return to the stock over the past six months? We interpret that as momentum trading. I was surprised to see it in the top five.

Another one that was interesting was also related to momentum. There was a paper that suggested to go long stocks closest to their 52-week highs, and go short stocks farthest from their 52-week highs. They get similar results. My results showed that stocks' 52-week high was the No. 4 most important variable for these professors.

SM: In your conclusion you wrote that what's important to these active investors is surprising and "somewhat troubling." Why troubling?

CW: Because the things these professors seem to be paying attention to are not the things we give attention to in the classroom. From a pedagogic perspective, maybe we're teaching the wrong stuff. But I understand why we do what we do we start with the traditional stuff, and we don't have time to focus on the more recent stuff. But that's the stuff that's actually relevant for buying and selling.

SM: Exactly how active are these investors?

CW: They're not as active as you might think. Considering what our responsibilities are, we thought maybe there's more practical experience in the sample. The median finance professor has bought a stock 10 to 19 times in their lifetimes, and 14.5% have never purchased a stock in their lives, which is also surprising.