Knowledge that Transforms

International Development

To most economists the single equation least squares regression model, like an old friend, is tried and true.Its properties and limitations have been extensively studied, documented and are, for the most part, well known.Any good text in econometrics can lay out the assumptions on which the model is based and provide a reasonably coherent --perhaps even a lucid -- discussion of problems that arise as particular assumptions are violated.A short bibliography of definitive papers on such classical problems as non -normality, heteroscedasticity, serial correlation, feedback, etc., completes the job.As with most old friends, however, the longer one knows least squares, the more one learns about it.An admiration for its robustness under departures from many assumptions is sure to grow.The admiration must be tempered, however, by an apprecia- tion of the model's sensitivity to certain other conditions.The requirement that independent variables be truly independent of one another is one of these.Proper treatment of the model's classical problems ordinarily involves two separate stages, detection and correction.The Durbin -Watson test for serial correlation, combined with Cochrane and Orcutt's suggested first differencing procedure, is an obvious example.*

Economists and other users of statistical methodology often posit a probabilistic model of some real world phenomenon which has more unknown parameters than there are sample observations. In such cases it is usually impossible to estimate jointly all parameters from the sample data. Even in those instances where there are well established estimation procedures when the number of sample observations, n, is than the number of parameters, r, these methods are generally inadequate when n < r, as the necessary calculations cannot be carried out. Furthermore, when n is only slightly larger than r, such estimates often prove to be unreliable in more than one sense. One particular example of such a problem that frequently occurs in analysis of psychological and economic data can be paraphrased as follows: the researcher posits a linear regression model as defined in equation (1) below with r 1 independent variables, but only n < r observations are available. Since the standard least squares procedure cannot be applied, he may ask the following seemingly reasonable question: What subset of the r 1 independent variables should I select for inclusion in a new model to which I can apply the standard least squares procedure? Our purpose here is to demonstrate that one frequently used ad hoc method for determining such a subset by ordering simple sample correlation coefficients can be highly misleading. Any procedure which uses a given set of sample data both to determine the structure of the model to be estimated and to estimate parameters of this model is intuitively unsettling. Here we present tables which quantitatively demonstrate how dangerous such ad hoc methods can be. II Ad Hoc Use of Simple Correlation Coefficients to Determine Model Structure

info:eu-repo/semantics/published

Wassily Leontief, Alison Morgan, Karen Polenske, David Simpson, Edward Tower, The Economic Impact--Industrial and Regional--Of An Arms Cut, The Review of Economics and Statistics, Vol. 47, No. 3 (Aug., 1965), pp. 217-241

Draft

W l THAT have we learned from ComIV Vmission on Money and Credit about processes and responses in monetary policy? So far as I can see, from Report itself, very little but perhaps experts in monetary economics were not a significant part of audience at whom Report was aimed. So most fruitful course may be to turn primarily to staff papers. Here there is substantial evidence of quantity theory reborn at least in sense that money matters a good deal in determining aggregate spending on current output and of an important postwar change in monetary theory. As noted in my introduction to this volume, this postwar shift in theory is centered in approach to demand for money and other assets. In this theory, channels through which variations in money supply affect levels of income and employment include not only a change in the interest but also changes in relative prices of all assets-real and financialwhich in turn lead to shifts in spending on existing assets and currently produced goods and services. Thus in its extreme form approach suggests that we need to look not at one interest rate but at an extremely large number, including implied interest rates on all real assets and including consumer goods of any degree of durability. In logical terms, this approach offers an elegant rapprochement for devotees of and quantity-theory approaches to role of money, and for monetary-versus-fiscal policy disputes since I930's. If look prevails, we may look back at much of this controversy as a good deal less significant than it has seemed en route. The great controversy over Is savings really equal to investment? of late I930's and early I940's springs to mind. But agreement on this mechanism doesn't necessarily tell us how important money is quantitatively. If we look at staff papers (or at least, at sample I managed), what does professional support for renaissance of money and monetary policy amount to? Friedman and Meiselman, as might be expected, plump for money as a prime determinant of level of spending on current output, and show convincingly that a simple, traditional monetary model versus a simple traditional Keynesian model test gives verdict clearly to stability for velocity over stability for ratio of autonomous to income, including cases where reasonable lags are introduced. Moreover, they go on to spell out portfolio balancing mechanism as at least a plausible mechanism through which this monetary effect may be exerted. If we take Section VI of their paper as a statement of new monetary orthodoxy, on intellectual grounds at least a good deal of basis for long quarrel between monetary and Keynesian economists has been reasoned (or compromised) away. Few, even most ardent neo-Keynesians, would disagree that impact of open market operations may be through spreading net which Friedman and Meiselman spell outnot merely through one (bond) interest rate alone acting on investment decisions. The major challenge to now generally accepted fiscal policy position as our really powerful stabilization tool becomes a strong one if a reasonably stable demand for money is added to mechanism as at least Friedman and Meiselman argue. The C.M.C. staff papers contain no empirical answer to Friedman and Meiselman challenge to show better results with another model. Tobin, in his paper on debt policy, provides an elegant statement of a very similar mechanism through which changes in money stock and liquidity may influence spending decisions on current output through rebalancing of asset portfolios. But I hope it will not be too dissident a note to suggest that we really know very little empirically about validity of this description of channels of monetary policy; and that elaborate portfolio-balancing general equilibrium approach lacks intuitive appeal

F what I have seen of the work of the Commission on Money and Credit, one thing stands out.It is a part of one question addressed by one task force to the Federal Reserve Board, and it reads as follows: "Given the customary credit control instruments and the ultimate objectives of price stability, highlevel employment, and economic growthhow is monetary policy formulated in the short run?For instance, what sort of factors are weighed in determining current policy, what guides are utilized and what are the immediate objectives of policy?"The merits of the Board's reply, the many contributed papers, the Commission's Report, and recent monetary policy itself, are all certainly debatablebut I submit, if you will pardon the use of strong language, that this is a damned good question.It asks, politely but firmly, that we omit the usual garbage about opposition to sin and advocacy of motherhood, and just say what it is we are trying to increase, decrease, or hold still.The Board's reply ran to some 35 doublespaced pages.The burden of drafting was carried by my colleague, Woody Thomas,