Knowledge that Transforms

SINCE KOYCK'S pioneering work, distributed lag models have come to play an important role in econometrics and much work has been done to develop methods for analyzing them-see e.g. Koyck [15], Klein [12], Solow [20], Fuller and Martin [5], Malinvaud [18], Hannan [9], Liviatan [17], Amemiya and Fuller [1], Zellner and Park [25], Thornber [21], Waud [23], Dhrymes [3], and Griliches [7]. In the present paper we present maximum likelihood and Bayesian estimation procedures for estimating the parameters of a typical distributed lag model under four alternative sets of assumptions regarding disturbance terms' properties. Then these procedures and assumptions are employed in analyses of a sample of United States quarterly consumption data to illustrate their application and to show the sensitivity of inferences to the assumptions made about disturbance terms' properties. We also compute posterior probabilities associated with four alternative models. The plan of the paper is as follows. In Section 2, the model to be analyzed is described and alternative assumptions about disturbance terms are introduced. Section 3 contains a discussion of maximum likelihood techniques and application of them in the analysis of U.S. quarterly consumption data. Then in Sections 4 and

J. Frank Sharp, James C. Snyder, James H. Greene, A Decomposition Algorithm for Solving the Multifacility Production-Transportation Problem with Nonlinear Production Costs, Econometrica, Vol. 38, No. 3 (May, 1970), pp. 490-506

In this paper four alternative quarterly econometric models of investment behavior are fitted to a common set of data for individual manufacturing industries in the United States. Goodness of fit and absence of autocorrelation of errors are used as a basis for comparison of the performance of the alternative models. The econometric models are compared with each other and with alternative explanations of data on investment based on surveys of anticipated investment and on mechanical forecasting schemes. The four econometric models included in our study are those of Anderson [2], Eisner [15], Jorgenson and Stephenson [38], and Meyer and Glauber [46]. On the basis of our comparison, the ranking of the alternative models is as follows: (1) Jorgenson-Stephenson, (2) Eisner, (3) Meyer-Glauber, (4) Anderson. Anticipatory data give a better fit to data on investment expenditures than that provided by any of the econometric models. Mechanical forecasting schemes provide a fit that is superior to the Anderson and Meyer-Glauber models. These schemes are slightly inferior to the Eisner model and clearly inferior to the Jorgenson-Stephenson model. The alternative econometric models included in our comparison differ in specification of the time structure of the investment process and in the role ascribed to specific determinants of investment behavior. Both aspects of an econometric model affect its performance so that it is difficult to discriminate among alternative determinants of investment behavior on the basis of our results.