Knowledge that Transforms

I wish to dedicate the paper to Mark Perlman, who guided this Journal until now. Perlman's help and encouragement in preparing this paper were characteristically vigorous and scholarly. I wish to express appreciation to the Hoover Institution where an early draft was written and to E. S. Shaw for his comments on that draft. Alex Cukierman, Brian Kantor, Scott Richard, and E. Roy Weintraub made several helpful suggestions, and Karl Brunner suffered through many discussions about Keynes and Keynesians. Many people read and commented on the previous draft, and their suggestions and criticisms have helped me to see points I would have missed. I am grateful especially to Paul Davidson and Donald Moggridge. Davidson commented generously and helpfully on almost every page. Moggridge helped me to strengthen my argument and graciously made available sections of volume 27 of Keynes's Collected Writings that had not been published at the time.

of economic time series similar to the observed high frequency series. A transformation that converts these series to serially uncorrelated stationary time series would therefore introduce the same conversion to the residuals. Although the discussion in this paper has been limited to the problem of distribution, a similar treatment can be given to the problem of interpolation and extrapolation by related series. As has been shown in the paper of Chow-Lin, the three problems can be treated simultaneously by properly defining the transformation matrix B.