The gross national product (GNP), a quarterly measure of the goods and services produced in the United States in constant dollars (dollars deflated for increases in prices), rose at an annual rate of 3.3 percent in third-quarter 1985. The White House had talked of a 5 percent rate in the second half of 1985, and of 1985 exceeding 1984 by 3 percent before the third-quarter report was made. News reports now say a 6.7 percent fourth-quarter annual rate will be necessary for a 5.0 percent second half.

What are all these figures and what do they mean? One thing is certain, no matter how you figure it, there will not be a 5 percent second-half annual-rate rise, nor will 1985 exceed '84 by 3 percent. The economy simply cannot get there from here.

For those of you who want to read further, I am sorry I have to take you through a statistical labyrinth. But if the country is going to face up to what needs to be done to achieve economic goals, it will take individuals who are not afraid to learn to distinguish between statistical nonsense and statistical make-sense.

To begin with, what is a quarterly annual rate of 3.3 percent? It is the result of compounding the actual quarterly percent rise four times. Thus, 0.82 percent rate in one quarter, if maintained for four quarters, would result in a four-quarter increase of about 3.3 percent.

Ironically, the 3.3 percent annual rate is a monthly rate of less than 0.3 percent, a rate that would be looked at as unusually mild if it had occurred in industrial production or housing starts. Yet, many observers greeted the 3.3 percent as a startling improvement over the second quarter's 1.9 percent annual rate rise. The actual difference between a quarterly increase of 0.8 percent and 0.5 percent is hardly earth-shattering.

Trying to achieve a 5 percent rise in the second half of 1985 would be earth-shattering. With GNP at an annual-rate level of $1,671.3 billion (or $1.671 trillion) in second-quarter 1985, GNP would have to increase in both the third and fourth quarters by a little more than 1.2 percent. It would mean a second quarter of $1,671.3 billion, a third quarter of $1,691.8 billion, and a fourth quarter of $1,712.6 billion.

But since the third-quarter ``flash'' estimate of $1,684.8 billion is $7 billion below the needed level, the fourth quarter would have to be $7 billion higher than we show, or $1,719.6 billion. This would require a 2.1 percent increase over the third-quarter $1,684.8 billion level -- or an annual-rate rise of 8.5 percent!

The 6.7 percent set forth by the news media (probably from a White House source) stems from adding the 3.3 percent third-quarter rise to 6.7 percent and dividing by 2 in order to obtain a 5 percent rise for the second half. While such arithmetic is simple and sweet, the method is simplistic and sour, totally inappropriate and inaccurate.

If you can keep holding onto your chair for a few more minutes, let's explain two more statistics.

You may have noted I wrote of an annual rate ``level'' in referring to a $1,684.8 billion GNP for third-quarter 1985. Now, the US economy did not produce $1,684.8 billion of goods and services in third-quarter 1985. It did not produce anywhere near that amount. The $1,684.8 billion is, in fact, four times the actual amount produced of $421.2 billion. Yes, four times what was produced!

Is somebody trying to statistic us to utter confusion? Not really, but statisticians forget that ordinary human beings read about these data or hear them on the radio or see them on TV. They forget that when a homemaker makes one pie in a week, he or she thinks one pie has been made, not 52 pies at an annual rate.

But for comparing how the economy is faring in the current 1985 quarter with what the 1984 annual level was or the '83 annual level, it is convenient to multiply the quarterly amount by four.

This brings us to our second statistic. Hold on tight.

The total amount of goods and services produced in 1984 was $1,639.3 billion. If 1985's GNP is to exceed '84's by 3 percent, it will have to be $1,688.5 billion (that's 1.03 times $1,639.3 billion).

If 1985 is to equal $1,688.5 billion, the amounts produced in the four quarters of '85 will have to total $1,688.5 billion. But if each quarter's figure is four times what was actually produced in the quarter, we have to add them up and divide by four to obtain the yearly amount produced.

With the annual rate-levels in the first three quarters of 1985 as follows -- first quarter, $1,663.5 billion; second, $1,671.3 billion; third, $1,684.9 billion -- then fourth-quarter GNP will have to be $1,734.4 billion, or an annual-rate rise of 12.3 percent (2.9 percent compounded for four quarters).

That's a fourth-quarter annual rate dollar increase of $49.6 billion, compared with increases in the first, second, and third quarters of $1.6 billion, $7.8 billion, and $13.5 billion, respectively.

There is just no way the economy can produce $49.6 billion more in the fourth quarter than in the third quarter. There just is no way GNP can increase at a 12.3 percent annual rate in the fourth quarter following annual-rate increases of 0.3 percent, 1.9 percent, and 3.3 percent in the first three quarters of the year.

There just is no way 1985 GNP can exceed 1984 by 3.0 percent. No way at all.

Mr. Lempert is director of Statistical Indicator Associates in North Egremont, Mass.