Lectures in econometric modelling (ed. 2015)

Maria Grazia Zoia, Mario Faliva, Mario Faliva, Mario Faliva, Mario Faliva

• Publisher: EDUCatt Università Cattolica
• Year: 2016
• ISBN:
  9788867809851
• Topic: Economia
• Subtopic: Econometria

This book collects a series of lectures on classical econometrics given, at different times, at the Catholic University of Milan and at the Geneva University, on the one hand, as well as within several PhD. programs in Economics and Statistics, on the other.
As prerequisites to successfully tackle the topics dealt with in the book, we recommend the usual toolkit of linear algebra and mathematical statistics, beyond a knowledge of the principles of econometrics.
The book is essentially meant to provide a thorough overview on the theory and methods of multi-equation econometric models.
The topics to be investigated move from a conspectus of forms of both linear and non-linear models, in static and dynamic frameworks, to causal structure analysis from a system-theoretic standpoint, to continue with parameter identification and estimation issues, and simulation and prediction techniques, to conclude with multiplier analysis and policy utilization of econometric models.
A useful insight into the book design and intents can be gained by looking at the table of contents in the light of the flow-diagram shown in the front cover.
Milan, February, 2014
Mario Faliva – Maria Grazia Zoia

CHAPTER 1
Formal Aspects of Econometric Models

1. Econometric Model Building Activities

As a rule, in any econometric research we may distinguish four stages, namely:

1. specification;

2. identification and estimation;

3. validation;

4. final uses.

The mechanism connecting the above mentioned step may be schematically presented as a flow-chart, i.e.:

2. The Structural Form

An econometric model takes generally the form of a system of equations involving a set of variables. The variables represent the quantities which are supposed essential for the problem under consideration and the equations formalize the relations among the variables suggested by the economic theory.

The endogenous variables, which are the quantities explained by the model under consideration, are the so called dependent variables. In the following such variables will be denoted by the symbol y. In multivariate models the endogenous variables enjoy the property to be jointly determined by the model, insofar as they represent its solution.

The exogenous variables are the quantities which are determined outside the economic system represented by the model; they are the so called independent variables and in the following they will be denoted with the symbol x.

As well known, there is an asymmetric relationship between endogenous and exogenous variables. The latter explain the former without depending on these ones.

Besides, it is worth recalling that the distinction between these two kinds of variables don’t depend on the nature of variables but only on the assumptions of the model specification, mainly on its dimension.

Econometric models should be – in order to be realistic- dynamic, stochastic and not necessarily linear, and so as a rule, specified in parametric form.

As a rule, econometric models are dynamic. This happens because impacts of explicative variables on the endogenous ones are not gener...

Table of contents

• INDICE
• Preface
• CHAPTER 1 Formal Aspects of Econometric Models
• CHAPTER 2 Causal and Interdependent Systems
• CHAPTER 3 From the Model to the Structure
• CHAPTER 4 The Model Validation
• CHAPTER 5 Econometric Prediction
• CHAPTER 6 The Use of Econometric Models for Policy Evaluations
• ADDENDUM An Approach to Linear Dynamic Models Bridging Classical and Time-Series Econometrics
• INDICE
• References