Economic Modelling

5 Key excerpts on "Economic Modelling"

• 

  eBook - PDF

  Models, Simulations, and the Reduction of Complexity

  • Ulrich Gähde, Stephan Hartmann, Jörn Henning Wolf(Authors)
  • 2013(Publication Date)
  • De Gruyter

    (Publisher)

  Each discipline or type of discipline has its own ways – styles, routines, values, and conventions – of modeling, and they are not a simple function of the specific nature of its target domain. It is always somewhat risky to make gener-alizing claims about a discipline and its characteristic practices and the values guiding those practices, but I believe the following will be recognized as more or less accurate regarding much of economic modeling in the recent decades. First , economic modeling is often theory-driven, shaped and constrained strongly by the dominant theoretical framework. This framework nowadays usually requires that models be built in terms of optimizing agents and equilib-rium outcomes. Second , the combination of parsimony and breadth is highly valued in eco-nomics. This means that there is an urge to increase the unification of diverse phenomena in terms of portable model structures or modeling principles, that is, structures or principles that can easily be transferred from one domain (or even discipline) to another (see Mäki 2001, 2009). Third , what is typically highly valued in economic modeling is mathematical rather than numerical precision. It is not surprising therefore that analytical deri-vation tends to be preferred to computer simulation (see Lehtinen and Kuorikoski 2007). Fourth , among the achievements of theoretical modeling economists often mention that models provide some “insight” into phenomena and the mecha-nisms which produce them; that they yield conditional predictions that state that if certain conditions were to prevail, then this or that would happen; and that they suggest how-possibly explanations that give account of ways in which some given phenomena might have come about (in contrast to how they actually did come about). Contested Modeling: The Case of Economics 89 Fifth , the so-called Duhem-Quine problem of underdetermination of theory or model choice by the empirical data is particularly pressing in economics.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Applying social science

  The role of social research in politics, policy and practice

  • David Byrne, Byrne, David(Authors)
  • 2011(Publication Date)
  • Policy Press

    (Publisher)

  SEVEN Modelling: representing the world in order to understand how it works

  Model as verb: To devise a (usually mathematical) model or simplified description of (a phenomenon, system, etc).

  Model as noun: A simplified or idealised description or conception of a particular system, situation, or process, often in mathematical terms, that is put forward as a basis for theoretical or empirical understanding, or for calculations, predictions, etc; a conceptual or mental representation of something. (OED)

  So models are representations of the world but are necessarily simplified. Something has to be left out. Bradley and Schaefer put it like this: ‘Modeling is the process of formalizing our framework for understanding the world around us by abstracting from a reality that is otherwise too complex for us to understand. In fact modeling is the central intellectual method that characterizes most empirical and mathematical approaches to the social sciences’ (1998, p 23, original emphasis).

  Immediately we have to identify a problem. If we are dealing with social reality composed of complex systems can we ever model by simplifying? If we leave anything out is it not the case that our representation – re-presentation, presentation again – will be inherently flawed because in complex systems everything matters in interaction with everything else and to omit anything means that we will have a model which cannot function to describe a system? So do we give up? Not at all. Let us return to Paul Cilliers’ wise remark already quoted in Chapter One:

  The most obvious conclusion drawn from this perspective is that there is no over-arching theory of complexity that allows us to ignore the contingent aspects of complex systems. If something is really complex, it cannot be adequately described by means of a simple theory. Engaging with complexity entails engagement with specific complex systems. Despite this we can, at least at a very basic level, make general remarks concerning the conditions for complex behaviour and the dynamics of complex systems. Furthermore, I suggest that complex systems can be modelled. (Cilliers, 1998, p ix)

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Models, Mathematics, and Methodology in Economic Explanation

  • Donald W. Katzner(Author)
  • 2017(Publication Date)
  • Cambridge University Press

    (Publisher)

  Economic models, as noted above, do not exist in reality though they themselves are, of course, “real” in the sense that their formal-logical description is stated and communicable. They are mental constructions that are quite distinct from, albeit hypothetical approximations to, the reality to which they relate. To confuse a model of a person, a firm, or an economy with, respectively, an actual person, an actual firm, or an actual economy, is to commit what Machlup called the “fallacy of misplaced concreteness.” According to Machlup [20, p. 9], the fallacy occurs when theoretical symbols are used “as though they had a direct, observable, concrete meaning.” What a model normally does do, of course, is to establish principles and modes of causation that can be applied only in very broad and general ways. Sixth, all models, in their construction, are subject to what may be called the “angle of vision” that the scholar who builds them brings to his work. Angles of vision emerge from pre-analytical persuasions arising out of backgrounds and experiences, and they influence, in turn, the nature of the questions asked and the assumption content of the analyses put forward to answer them. In this way, and in a very broad sense, all scholarly enterprise, in particular model building, is inescapably contaminated by ideology, politics, culture, and values. 13 The economic phenomena that explanatory economic models purport to address largely consist of economic behavior or the consequences of that behavior. The models themselves can be constructed with one or more explanatory objectives in mind. One possible objective could be merely to center attention on describing the manner in which an actual economic entity may be structured. Such a construction could be thought to clarify the characteristics of the phenomenon in question and explain how its various possible parts might cohere.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Econometric Modeling

  A Likelihood Approach

  • David F. Hendry, Bent Nielsen(Authors)
  • 2012(Publication Date)
  • Princeton University Press

    (Publisher)

  Chapter Eleven Empirical models and modeling In the development so far, we have been concerned with statistical models for cross-sectional data. In the subsequent Chapters 12 to 17, in a similar way we will develop and analyze statistical models for time-series data. First, however, it is useful to consider some broader aspects of econometric modeling. Initially, we will discuss the main motivation for econometric modeling, leading to a definition of empirical models. We can then proceed to a more detailed analysis of the possible interpretations of regression models. Finally, the notions congruence and encompassing are introduced to describe the aims of econometric modeling. 11.1 ASPECTS OF ECONOMETRIC MODELING To account for the empirical evidence obtained in economics, economists treat ob-served data outcomes as the realizations of random variables that have been gen-erated by economic behavior. The distribution function F ( x ) = P ( X ≤ x ) fully characterizes the properties of the random variable X , but F is typically unknown. Consequently, we formulate statistical models, which are families of possible dis-tributions, indexed by a vector parameter θ . In the context of economic theory, we consider particular distributions, rather than families of distributions, with para-meters denoted θ 0 that arise from the solutions of decision problems by economic agents: e.g., the marginal propensity to save (about 0 . 1 on average) is selected by agents when maximizing lifetime utility. As discussed in previous chapters, esti-mators of unknown parameters can be based on the observed data, derived from the likelihood function L X ( θ ) . Because data are realizations of random variables, such estimators have sampling distributions (so take different values in different samples), as do test statistics for hypotheses relevant to economics.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  The Microeconomics of Public Policy Analysis

  • Lee S. Friedman(Author)
  • 2017(Publication Date)
  • Princeton University Press

    (Publisher)

  First, policy conclusions are often quite sensitive to varia-tions in the way the policy itself is modeled. Therefore, the analyst must take care to under-stand the details of any specific proposal before deciding how to model it (or attempting to evaluate another’s model of it). Second, the reexamination of assumptions that are stan-dard and appropriate in many contexts often becomes the central focus in a particular policy context. Indeed it is their inappropriateness in particular contexts that, in the aggregate, helps us to understand the large and varied roles of public policy in an economy such as those described in the previous chapter. Let us mention one final general modeling issue: the form that the model takes. Examples were given earlier of different forms that models might take: for example, the plastic model of an airplane and the mathematical model of gravity’s effects. We will sometimes con-struct economic models from verbal descriptions, sometimes by geometric representation, and sometimes by mathematical equations. Models may appear in the form of short stories or other abstractions. What should determine the form that the model builder chooses? In the different forms of economic models mentioned, the essence of the underlying behavior can be the same. What purpose is served by presenting it in different ways? We wish to note this distinction: Modeling is a way for the model builder to learn, but it is also a way to communicate with (or to teach) others. The main point of doing policy analysis is to learn: The analyst does not know the conclusion at the start and seeks to come to one by a logical procedure that can be subjected to evaluation by professional standards. Yet the form of a model used for learning is not necessarily appropriate as a form for com-24 Chapter Two munication. The latter depends upon the audience. If policy analysis is to influence policy, it is particularly important that it be communicated effectively.

  Sign up to read

  Learn more about book