No label defined (Q4947391)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1432780
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 1432780

      Statements

      0 references
      0 references
      17 April 2000
      0 references
      aggregation of time series
      0 references
      common components
      0 references
      Granger causality
      0 references
      microeconomic parameters
      0 references
      VAR
      0 references
      Wold decomposition
      0 references
      This book draws on the recent work of its authors in the area of aggregation of time series. The field developed around the seminal contribution of Clive Granger (1980) who showed that the summation of a large number of (possibly independent) ARMA (Autoregressive, Moving Average) time series yields a fractionally integrated process. The significance of this result is that the individual ARMA processes may display very little persistence while the aggregate (fractionally integrated) process exhibits long-memory. The interest of macroeconomists in this result derives from the observed persistence of unemployment, output and other macroeconomic time series. One of the leading theories of economic business cycles is the Real Business Cycle (RBC) theory originally developed by Robert Lucas, Ed Prescott and Thomas Sargent in the mid 70's. Unlike previous economic analysis of the business cycle, which viewed the volatility of output and employment as prima facie evidence of the economy not being at an equilibrium, the RBC literature develops the idea that shocks to technology can generate volatility in output and other economic variables. One problem associated with this explanation is that, in order to generate variability in output consistent with economic data, one needs to assume that technology shocks are themselves very persistent. It has been hoped that economic behavior would transform the productivity shock process with little persistence into a persistent output process for aggregate variable, but simulation work starting in the late 80's showed this not be the case, and this feature was dubbed ``the weak transmission mechanism of RBC''. The RBC literature had traditionally worked with representative agent models. Starting in the early 90's, econometricians started using heterogeneous process with little persistence to represent heterogeneous agents and used Granger's result to argue that the aggregation of these processes would exhibit persistence, a result that was needed if the RBC approach was to survive. The aggregation literature did not, until very recently, make a major impact on macroeconomics. The main problem that held back the field was that it was attempting to explain a major economic phenomenon by linearly aggregating time series. Very little attention was initially paid to what these time series ought to be, why they should be aggregate and how the individual decisions would adjust to this aggregation. In a nutshell, aggregation was seen as one of the many an ad-hoc explanations to the phenomena of persistence with no economic content. Since the publication of this book, the field has undertaken a major renovation and the standard for current work in the area is to start from solid macroeconomic foundations. This book is a good example of the state of aggregation in the mid-90's. Most of the material presented is econometrics in nature with only passing references to economic relationships and none at all to concepts of economic equilibrium or to the interactions between markets (labor market, capital market, goods markets, financial markets) that are so important to macroeconomic theory. An example of the approach of the authors is the importance granted to the so-called ``Alternative Principle'' which states that the null set of an analytical function over an open connected set is either of measure zero or the space itself. The authors feel that this principle has major implications for the relationship between microvariables and their aggregation. This book is not intended as a layman's introduction to the aggregation field either as it uses the trade's vocabulary without prior introduction or definition. This book would be most useful to readers with a solid background in the econometrics of time-series and who want a reference to the early works of the ``Italian School of Aggregation''. In part I of the book, the authors show that, in their framework, the effect of the idiosyncratic shocks affecting individual consumers is overwhelmed by the effect of the common shocks. Therefore, only the common components survive aggregation. They also argue that US data support the hypothesis of multiple common shocks. Part II builds the aggregation framework from individual agents' problem and states the Alternative Principle. This principle is then applied to the rank of the vector of aggregation, to cointegration, Granger causality and VAR and ARMAX analysis. They show that, generically, a property that holds for an individual problem does not hold for the aggregate problem. Part III presents applications to the problem of consumption smoothing (the question of ``excess sensitivity of consumption'') and to the question of disaggregating of the business cycle.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references