Content area
Full Text
Summary
Progress, regress, and continuity in quantitative analysis for policy are discussed. We look at the present from the perspective of Tinbergen, Haavelmo, and Keynes. Probability modeling has been in retreat at central banks and elsewhere. New computational methods, though, are making Bayesian analysis of previously intractable problems possible, and at the same time appreciation of the clarity with which Bayesian data analyis integrates with decison-making is spreading.
Key words: econometrics, methodology, policy
A few years ago I wrote 'Macroeconomics and Methodology' (1996). While I will not rehash here the views set forth in that paper, those views are the starting point for this paper. This paper discusses and draws connections among three sources:
1. Visits to central banks and interviews I conducted with people in those banks in 2002, assessing the models they use and how they use them; 1.
2. Recent technical developments that have converted theoretical advantages of Bayesian over classical approaches to inference into practical reality in some applied areas; associated applied work and methodological commentary emerging in the literature;
3. Haavelmo's 1944 paper/monograph 'The Probability Approach in Econometrics', and some related previous literature.
The paper begins by discussing 3, using it as a kind of table of contents for aspects of 1 and 2.
1 THE ROAD AHEAD, AS OF 1939, AND OUR PROGRESS ON IT
To undertake what Tinbergen did in the 30's, constructing a quantitative model of the economy based on tools of statistics then available, required audacity, energy, and a certain personality type. This personality type was not that of Keynes. He closes his review of the first volume of Tinbergen's League of Nations modeling effort with
I hope I have not done injustice to a brave pioneer effort. The labour it involved must have been enormous. The book is full of intelligence, ingenuity, and candour; I leave it with sentiments of respect for the author. But it has been a nightmare to live with, and I fancy that other readers will find the same. I have a feeling that Prof. Tinbergen may agree with much of my comment, but that his reaction will be to engage another ten computors and drown his sorrows in arithmetic.
(Of course, in those days, what Keynes meant by 'computors'...