Gerhard Tintner, Econometrics

Gerhard Tintner’s Econometrics, first published in 1952 and represented here by a reprint of undocumented date, is a textbook organized around the statistical problems of economic investigation. Its progression—from methodological foundations and policy applications through multivariate and simultaneous-equation methods to time-series analysis—makes the choice of analytical procedure central to economic interpretation. An appendix on matrix theory and numerical procedures supports this practical orientation.

Tintner’s methodological emphasis is explicit:

The main difficulties in econometrics are still statistical.

The book accordingly treats econometrics as a discipline in which economic questions must be matched with defensible methods of estimation. Its discussion of applications also draws a boundary between empirical inquiry and the ends that policy should pursue:

The choice of policy, however, is a matter of politics, ethics, and similar considerations.

Statistical analysis can inform policy without determining its normative objectives. This distinction gives the introductory discussion a relevance beyond technique: the authority of quantitative investigation does not extend automatically to political or ethical choice.

The treatment of multivariate analysis and simultaneous equations develops a corresponding caution about inference. In estimating price–quantity relationships, an observed association cannot simply be read as the economic relation sought:

The simple regression of the price on the quantity, or of the quantity on the price, cannot in general supply these estimates.

The conceptual move is from fitting an isolated relationship to considering a system in which variables are jointly determined. Reversing the direction of regression does not resolve that problem. The organization of the textbook thus connects the statistical difficulty announced at the outset with the economic structure that estimation must respect.

The final part examines economic observations through time. Chapters on trend removal, oscillatory and periodic movements, dependence between successive observations, and transformations address both the interpretation of temporal patterns and the consequences of manipulating data. Tintner questions a rigid separation of long-run movement from cyclical fluctuation:

But it seems that both trend and cycle ought to be explained by the same stochastic mechanism with economic time series.

Trend and cycle become aspects of a common explanatory problem rather than necessarily independent components. This position also limits the claims that can be made for a convenient fitted curve:

Polynomial trends should never be used for extrapolation.

The warning distinguishes describing past observations from justifying a forecast. A related concern governs the discussion of smoothing:

The fact that the application of moving averages introduces autocorrelations and serial correlations into pure random series is not surprising.

Analytical procedures can generate the very patterns an investigator might otherwise attribute to economic processes. Across its treatment of policy, simultaneous determination, and time series, the textbook therefore insists on distinguishing what a technique produces from what the evidence warrants. Its lasting methodological relevance lies in these linked cautions: quantitative results require an account of economic structure, statistical dependence, and the limits of inference.