Skip to main content

2023 | Book

Forecasting from Multi-equation Econometric Micromodels


About this book

Forecasting from multi-equation models has very rarely been the focus in econometric literature. In response, this book presents a range of methodologies to approach this complex field and offers readers essential information on forecasting from multi-equation econometric micromodels.

In the twentieth century, significant interest in econometric macromodels emerged. These multi-equation models are mostly systems of interdependent equations, most often used to describe the national economies of various countries. The book analyzes econometric forecasting procedures and illustrates them with empirical examples that are based on real economic (mostly business-derived) data. The procedure of forecast building from systems of interdependent equations is presented for two categories of econometric models: models with a feedback effect and models with closed-loop links between interdependent variables.

The forecasts obtained via this technique are compared with the results derived from reduced-form equations of the respective econometric model. The author also generalizes the rules of the reduced-recursive (helical, iterative) procedure application, against the backdrop of the proposed method of forecast building from reduced-form equations of systems of interdependent equations. Given its scope, the book will appeal not only to PhD students and researchers, but also undergraduate students and academics in general.

Table of Contents

Chapter 1. Single-Equation Econometric Model
The first chapter is devoted to econometric models in the form of one stochastic equation. Linear, product, and limited dependent variable models were presented. In this chapter all stages of building a single-equation econometric model are discussed. These are: specification, estimation of parameters, verification, and exploitation possibilities. Basic methods of estimation of structural parameters and parameters of the stochastic structure of the model are presented. Basic statistical measures determining the quality of an empirical econometric model are also discussed.
Jerzy Witold Wiśniewski
Chapter 2. Multi-Equation Econometric Models
The second chapter presents elementary information connected with the construction of econometric models consisting of many equations. Such models are used to describe economic systems or only parts of them. Three classes of these models are presented, i.e. simple models, recursive models, and systems of interdependent equations. The structural form of such a model and its reduced form are discussed. The essence and the need to identify the multi-equation model were explained. The basic methods of estimating the parameters of models belonging to each of the indicated classes are also characterized.
Jerzy Witold Wiśniewski
Chapter 3. Econometric Forecasts
The subject of the third chapter are the basics of econometric forecasting. Basic concepts used in econometric forecasting are presented. The basic assumptions of the econometric prediction theory and the consequences they are not met are discussed. The procedure of building forecasts based on a one-equation model was characterized. The issues of admissibility of forecasts along with their measures were explained. The issues of testing the accuracy of forecasts were discussed. The specificity of building forecasts based on multi-equation models was presented—the differences in forecasting procedures based on simple, recursive models and systems of interdependent equations were indicated.
Jerzy Witold Wiśniewski
Chapter 4. Forecasting from Simple Econometric Micromodels
Three empirical simple models based on real data from enterprises were presented. The first model describes the cost volatility mechanism of a company listed on the stock exchange. Forecasts of production, sales, and management costs were built, observing the rule of sequential forecasting. Another simple model was based on cross-sectional data. It describes the influence of personal characteristics of workers on their individual work efficiency and quality of production. The use of an empirical model to predict the performance and quality of candidates for a vacant worker position is presented. The third of the empirical simple models, also based on cross-sectional data, describes the mechanism of influence of the personal characteristics of traders on sales effectiveness and the effectiveness of debt collection. Forecasts of the variables described in this model and their use in the decision to employ candidates for the vacant position of a trader in the enterprise are presented.
Jerzy Witold Wiśniewski
Chapter 5. Forecasting from Recursive Econometric Micromodels
Three empirical recursive multi-equation models are presented. The chain procedure of building forecasts from the medium-sized enterprise model, which consists of seven stochastic equations, is presented. Chain prediction alternated with sequential prediction. The second recursive model consists of three stochastic equations. It describes the costs of an enterprise selling sports equipment that is listed on the stock exchange. The procedure of building quarterly forecasts together with the assessment of their admissibility was presented. The third recursive model characterizes the payment card market in China. The links between the most important variables in the Chinese model of the payment card market were characterized. The procedure for building quarterly forecasts for this market, described by seven endogenous variables, is also presented.
Jerzy Witold Wiśniewski
Chapter 6. Forecasting from an Econometric Micromodel in the Form of a System of Interdependent Equations
The sixth chapter discusses the specificity of forecasting a small enterprise as an economic system. Feedback and closed cycles of connections appear in such a system. This means that the econometric model is a system of interdependent equations. The classical approach to predicting from a system of interdependent equations is to use the empirical equations of the reduced form. This paper presents an iterative forecasting method based on the structural form of the system of interdependent equations, which guarantees synchronization of forecasts. Therefore, in the first place, an iterative forecasting method based on the econometric micromodel of an enterprise with a closed cycle of relationships was presented, assuming the inertia of the system. The next step were the forecasts, obtained iteratively, while interfering in the system with the use of control variables. The last prognostic solution for the system with feedback was the forecasting of financial liquidity and the effectiveness of debt collection using earlier forecasts of variables that formed the cycle. Forecasts obtained by the iterative method guarantee synchronization in the system, while forecasts from the reduced form based on the econometric micromodel result in discrepancies, i.e. lack of synchronization.
Jerzy Witold Wiśniewski
Forecasting from Multi-equation Econometric Micromodels
Jerzy Witold Wiśniewski
Copyright Year
Electronic ISBN
Print ISBN

Premium Partner