Financial analysts make more accurate price targets for the stocks they follow if they use more rigorous valuation models, according to a study from the University of Iowa.

The study found portfolios from analysts who used the complex Residual Income Model (RIM) saw a mean increase of 15 percent in one year, while portfolios from analysts who used the simpler Price/Earnings Growth (PEG) ratio decreased in value by better than 5 percent.

“It doesn’t matter how good the analyst is at other forecasting tasks, a less sophisticated model is less successful at forecasting prices than more sophisticated models,” says study co-author Cristi Gleason, professor of accounting in the Tippie College of Business.

The research is important because millions of investors decide what stock to buy based in part on what analysts are predicting the stock price will be in the future. Analysts arrive at these price target forecasts using a variety of different models that incorporate different financial metrics and statistics, such as stock price, forecasts of future earnings, and projected growth rates. But more complex models use factors that, for instance, incorporate the current state of the firm, and more explicitly measure risk, factors that simpler models like the PEG do not consider.

However, many earlier studies have shown these price targets are not accurate. In their own study, the UI researchers found that in a pool of 45,600 price targets, the analysts’ forecast a mean return of 82 percent. “That means they expect stock prices to almost double,” Gleason says. “That doesn’t happen very often. They’re a little over-optimistic.”

But in their study, Gleason and her co-researchers found that the analysts using simple models were far less accurate than those using complex models. With her co-authors Bruce Johnson, professor of accounting in the Tippie College of Business and Haidan Li of Santa Clara University, Gleason looked at 21,200 forecasts made by equity analysts between 1997 and 2003. They put those forecasts into six portfolios, based in part on how the analysts rated the stock—strong buy, strong sell, for example.

They inferred whether the analysts were using the complex RIM or the simple PEG ratio from the numbers the analysts themselves provide in their reports. They used the analysts’ own earnings and growth forecasts in both the RIM and the PEG models. Comparing their results to the analyst’s result, they were able to determine whether the model used was more likely to be complex or simple.

The portfolio that most definitely used the complex model saw an increase in value of 15 percent in one year, and the portfolio performance grew worse as the model became simpler. From 15, it dropped to a 1.4 percent return, 6.94 percent, -3.52 percent, 2.97 percent, and finally -5.44.

“The research shows clearly that analysts who use complex models have better stock performance than those who use simple models or ratios,” Gleason says.

Gleason says the complex ratios were more accurate because they more rigorously examined more data, providing a truer picture of the firm’s stock.

“If simple ratios were just as accurate as complex models, then we should just teach the simple ratios in business schools and save everyone’s time,” says Gleason. “But this shows the more rigorous complex models we teach are more accurate, and are better for investors.”

The paper, “Valuation Model Use and the Price Target Performance of Sell-Side Equity Analysts,” was published in the Contemporary Accounting Research.