Want to get to grips with econometrics? Look no further. John F. Geweke (Iowa), Joel L. Horowitz (Northwestern) and Hashem Pesaran (Cambridge) offer Econometrics: A Bird’s Eye View (PDF here) in a new IZA discussion paper.
Prepared for the forthcoming New Palgrave Dictionary of Economics and Law, it provides a comprehensive 72 page overview of the subject, including a boggling 386 references! Here is the abstract:
As a unified discipline, econometrics is still relatively young and has been transforming and expanding very rapidly over the past few decades. Major advances have taken place in the analysis of cross sectional data by means of semi-parametric and non-parametric techniques. Heterogeneity of economic relations across individuals, firms and industries is increasingly acknowledged and attempts have been made to take them into account either by integrating out their effects or by modeling the sources of heterogeneity when suitable panel data exists.
The counterfactual considerations that underlie policy analysis and treatment evaluation have been given a more satisfactory foundation. New time series econometric techniques have been developed and employed extensively in the areas of macroeconometrics and finance. Non-linear econometric techniques are used increasingly in the analysis of cross section and time series observations.
Applications of Bayesian techniques to econometric problems have been given new impetus largely thanks to advances in computer power and computational techniques. The use of Bayesian techniques have in turn provided the investigators with a unifying framework where the tasks of forecasting, decision making, model evaluation and learning can be considered as parts of the same interactive and iterative process; thus paving the way for establishing the foundation of “real time econometrics”. This paper attempts to provide an overview of some of these developments.
The authors conclude that econometrics "has come a long way over a relatively short period", and that "both in theory and practice ..[it] has already gone well beyond what its founders envisaged." But there are also a number of difficulties. First, economic statistics are seldom "the results of designed experiments, but are obtained as by-products of business and government activities". They don't always suit economic analysis. Second, models are rarely fully specified:
..the specification of econometric models inevitably involves important auxiliary assumptions about functional forms, dynamic specifications, latent variables, etc. with respect to which economic theory is silent or gives only an incomplete guide.
Third, economic theory is incomplete and:
..on its own cannot be expected to provide a complete model specification has important consequences for testing and evaluation of economic theories, for forecasting and real time decision making. The incompleteness of economic theories makes the task of testing them a formidable undertaking.
In general it will not be possible to say whether the results of the statistical tests have a bearing on the economic theory or the auxiliary assumptions. This ambiguity in testing theories, known as the Duhem-Quine thesis, is not confined to econometrics... The problem is, however, especially serious in econometrics because theory is far less developed in economics than it is in the natural sciences.
Plenty of food for thought here.