With just about every economic forecaster trying to figure out whether the country is in a recession, Federal Reserve Chairman Alan Greenspan suggested yesterday that their task is becoming more difficult because "the structure of the economy is more like a moving target than a sitting duck."

"While our forecasting tools have improved considerably over the postwar period, our forecast accuracy has not," Greenspan told hundreds of members of the National Association of Business Economists meeting here.

Forecasters do not do a better job, the Fed chairman said, because they are dogged constantly by two key difficulties: They usually do not know where the economy stands even as they try to say where it is headed; and their predictions are based on the economy's past behavior, which may have changed without their knowing it.

"We are engaged in a continual struggle in which the benefits of improved techniques are eroded by an increasingly elusive and complex economic structure," said the Fed chairman, who told Congress last week that the recent sharp run-up in oil prices is making the "short-term outlook {for the economy} ... weaker."

The Fed tries to deal with these problems by using information provided by a mathematical "model" of the economy, modified by a healthy dose of judgment. The model -- basically a mathematical description of how various parts of the economy relate to each other -- provides "a useful starting point," Greenspan said. The problem is that models can never work very well because what they measure is constantly changing.

But instead of accepting a model's forecast, he said, the central bank relies heavily on the economic intuition of its policy makers and staff.

"Basically, he was saying there is no substitute for good old-fashioned judgment," said NABE President James F. Smith, a finance professor at the University of North Carolina who formerly was chief economist for Union Carbide Corp. "Models alone won't do."

Smith, himself a forecaster for years, added that no matter how much uncertainty surrounds a forecast, business executives as well as government policy makers like Greenspan don't have the luxury of waiting for several months or years to see if a forecast is right.

"They can't wait for two years for the data to be refined," Smith declared. "They have got to sit there and decide when to pull the trigger."

So, even though they may produce an imperfect product, forecasters stay in business because imperfect information is better than none at all. And most economists are careful to acknowledge their uncertainties so their clients can take them into account.

For example, in explaining the latest forecast from the WEFA Group of Bala Cynwyd, Pa., one of the best-known forecasting firms in the country, economists Larry Kimbell and Glenn Forman wrote:

"It is worth noting that such clear signs {of the beginning of a recession} typically do not manifest themselves until several months after a downturn has begun. Bearing this in mind, we continue to interpret the recent data as painting a picture of a very weak economy."

Greenspan knows the difficulties of forecasting as well as anyone because he was in the business for three decades before becoming Fed chairman. At the central bank he still pores over statistics as he gauges the pulse of the economy.

"Economic forecasting is really the art of identifying tensions in the economic process and understanding in what manner they will be resolved over the short to intermediate term," he said in his speech.

For Greenspan, a "tension" develops when one piece of the economic puzzle changes, such as consumer spending going up. That could lead to higher prices, increased production and bigger business inventories, but initially a forecaster will not know which will change, or by how much.

"Clearly, detecting key imbalances is a crucial element in the forecast process and is one reason why determining where the economy is now is so important in assessing where it may be headed," Greenspan said.