from Wikipedia, the free encyclopedia

The Econometrics is a branch of economics that the economic theory and mathematical methods and statistical to bringing together data on economic theory models empirically to review and analyze economic phenomena quantitatively.


The term econometrics was developed by Ragnar Frisch and Joseph Schumpeter in the early 1930s. This process culminated in the establishment of the Econometric Society and the Journal Econometrica (1933).

Econometric methods are mainly used in economics , but also in business administration . Regression analysis is one of the most important tools for estimating econometric models . In modern econometrics, computer-oriented models are becoming increasingly important. These include B. Monte Carlo simulation , nonlinear models , Bayesian econometrics , volatility , or artificial neural networks . Methods for exploiting quasi-experiments , for example instrument variables or regression discontinuity analyzes, are also widespread . In connection with panel data , linear panel data models and dynamic panel data models should also be mentioned.


Single equation

With individual equation estimates, only individual relationships are estimated:

The variable to the left of the equal sign ( e.g. consumption) is the variable to be explained , the explanandum, the variables to the right of the equal sign ( e.g. income) are the explanatory variables, the explanans. They are the constant quantities to be estimated, called regression parameters, and reflect the strength of the influence of the explanatory variable on the variable to be explained. A regression parameter to be estimated in this way would be the marginal consumption rate .

In the individual equations, the variables to be explained are also referred to as endogenous variables , the explanatory variables as exogenous . Often the variable to be explained appears to the left of the equals sign itself with a time delay as well as an explanatory variable on the right, the so-called delayed endogenous .


Econometric models consist of estimating equations and defining equations . Estimating equations such as the consumption function or the investment function are estimated with the help of econometric methods ( linear regression ). The estimation equations themselves are specified according to economic theoretical hypotheses (economic models), for example the hypothesis that consumption is determined by income or that investments change with the level of the interest rate.

Definition equations determine a fixed relationship between different features and are not estimated. So z. B. the gross domestic product by definition made up of consumption , investment , government expenditure and exports minus imports .

In contrast to the individual equations, the endogenous variables themselves appear as explanatory variables in other estimation equations. However, every econometric model also has exogenous variables that are not explained by any equation. These represent the assumptions of an econometric model. For example, an econometric model could leave the development of world trade unexplained and specify it as an exogenous variable (which then explains exports, etc.). Typical exogenous variables are also the tax rates, such as the sales tax rate.

If many variables of a model are exogenous, one speaks of a high degree of conditionality. If most of the variables explain each other endogenously, one speaks of a low degree of conditionality.

Additional Information

Not all information can be easily entered into a model mechanism. Such one-off influences have to be entered into the model "by hand" by changing the variable to be explained additively by a certain constant ("add") or by a certain factor ( add-factoring ).


If economic models are specified econometrically, i.e. if the mutual dependencies are estimated with the help of econometric methods, one obtains econometric (macro) models . This application of econometrics is called macroeconometrics . For example, an econometric model is used in the semi-annual joint diagnosis in Germany .

Such macroeconomic econometric models can be used to predict economic development ( economic forecast ) or to simulate economic policy measures. For example, it can be calculated how the increase in the VAT rate will affect the overall economy. However, it is controversial whether this can be recorded by such a model and whether econometric methods are superior to alternative methods.

In 1974, IBM developed an econometric planning language based on APL .

Alfred Nobel Memorial Prize for Economics

A prize donated in 1969 by the Swedish Reichsbank on the occasion of its 300th anniversary and considered the most prestigious in the field of economics. Since it is awarded annually together with the Nobel Prizes and is endowed with the same prize money, it is commonly referred to as the Nobel Prize for Economics or the Nobel Prize in Economics


Web links

Wiktionary: Econometrics  - explanations of meanings, word origins, synonyms, translations

Individual evidence

  1. ^ F. Schober and K. Spielberg: APL Econometric Planning Language (EPLAN). In: Proceedings of the eighth international conference on APL. Pp. 359-387 (September 22-24, 1976, Ottawa)