Murray L. Weidenbaum, departing tired and apparently frustrated from what he termed his "arduous task" as chairman of the Council of Economic Advisers, hinted yesterday that changes are needed in the conduct of administration economic policies.

Weidenbaum quoted John Maynard Keynes, the famous British economist, who wrote that policy makers need to be optimistic, but that when the optimism proves unfounded, "then we've got to try something else."

Weidenbaum spoke after President Reagan startled White House aides in St. Louis Thursday night by casually mentioning to a St. Louis television reporter that he had accepted Weidenbaum's resignation. The White House insisted that no policy dispute had prompted the action.

At a morning news conference, the CEA chairman, who will remain in his post until the end of August, declined to say what specific policy changes he would make in response to an economy that is not recovering from the recession as expected.

Instead, he underscored his support for Reagan's policy of cutting taxes and spending, restraining money growth and reducing regulatory burdens, saying, "I think the four pillars of the program are just right."

But then he hinted broadly that he favors some changes by drawing a distinction between basic economic policy and "the conduct" of a policy.

Yesterday afternoon, Treasury Secretary Donald T. Regan told reporters that the administration is reexamining its view of what is going on in the economy, but that no policy changes are in the works.

"We are working on the assumption that our policies are correct to take us out of the recession and keep down inflation," Regan said. "The signs seem to indicate that is what is happening . . . .

"If you are looking for a hint whether we are going to make a change before November, I think it is exceedingly doubtful," Regan continued. "The president doesn't want to do it. I don't want to do it. Unless there are major changes in these economic signs, we will not be making changes in this policy."

Regan defined a major change as, perhaps, inflation continuing to run at a 13 percent rate, "or the economic growth simply wasn't there, or we were growing much faster" than the 4 to 5 percent rate predicted earlier.

At his news conference, Weidenbaum indicated that such rates of growth in the gross national product, adjusted for inflation, will be one of the assumptions on which the administration is basing its mid-year budget review, due for release Monday.

"I think those numbers are well within the range of feasibility, possibility," he said. But he added, "I am not prepared to say they are the most likely forecast I personally would make."

Emphasizing his distinction between basic policy and how it is implemented, Weidenbaum noted that "there is a good deal of judgment involved in the conduct of economic policy."

Citing Keynes, he added that an economic policy maker needs "to take an optimistic and positive view of the future. And then Keynes goes on to say that if it doesn't work, then we've got to try something else."

By all accounts, Reagan's lack of flexibility in implementing his basic policy was a major factor in Weidenbaum's departure. Weidenbaum would not respond directly yesterday to questions about his inability to influence the president on numerous occasions, but emphasized that he had received full access to the president.

"I am confident that at every key stage of the process the president knew where the chairman of the Council of Economic Advisers stood," Weidenbaum said.

While he decided to resign last May, effective in time to return to the teaching post at Washington University in St. Louis from which he is on leave, Weidenbaum said he had not told the president or anyone else in the administration. Had he done so, he said, the news would have leaked out.

"I didn't want to become a lame duck," he said.

The resignation was not in response to any single, specific disagreement over policy, such as the nature of the economic assumptions in the forthcoming mid-year review, the CEA chairman indicated.

He said he was a party to the decision simply to extend the economic assumptions contained in the current congressional budget resolution rather than prepare a new forecast that would be at odds with those assumptions. "I see great value in using the same set of assumptions that Congress has used," he said.

The timing of his resignation, Weidenbaum said, was tied to the fact that work on that review would be complete, that the academic year begins in September and that the CEA's work on the next economic report would not begin until fall.

He also made it plain that he is physically tired from his daily grind. The CEA chairmanship does not have the support staff of a Cabinet secretary and represents "an arduous task," he said.

Furthermore, he said he had not paced himself with an eye to staying a full four years, something only his immediate predecessor, Charles L. Schultze, had done.

"I put in those 16 to 18 hour days as frequently as I could . . . . I did not stretch out my energy," he said.

Regan, who said he had not known of Weidenbaum's plans to leave, said the administration is reexamining its views on the state of the economy because of revisions in the gross national product statistics reported by the Commerce Department this week.

" . . . The downturn in the fourth quarter of 1981 and the first quarter of '82 was much deeper and sharper than we had expected," Regan said.

Instead of a drop in the output of goods and services at a 4.5 percent seasonally adjusted annual rate in the fourth quarter and at a 3.7 percent rate in the first, the revisions show declines of 5.3 percent and 5.1 percent, respectively. "We have to take a look at that," he said.