Simultaneous equation models:
The econometric models we studied so far talk about single equation regression models where a dependent variable is related to a set of independent or explanatory variables. In all these regression models, y is the 'dependent' or 'explained' variable arid x1, x2,.... are the 'independent' or 'explanatory' variables. Thus y depends upon x, but the reverse is not true. In real life, however, we normally find variables that are dependent on each other. In this sort of relationship the equilibrium values of the variables can be determined by considering the simultaneity in relationship.
In simultaneous relationships a single equation under investigation is part of a wider phenomenon. In such cases the variables can be explained with the help of a system of equations. Let us take an example. At the macro level, as you know from macroeconomics course, aggregate consumption expenditure epends upon aggregate disposable income. Aggregate disposable income, on the other hand, depends upon the national income and taws imposed by the government. Moreover, national income is dependent on aggregate consumption expenditure of the economy.
Thus estimation of a single equation between aggregate consumption and disposable income will not provide us with consistent and unbiased estimators.
Unlike the single equation model, in simultaneous equation ilodels we get more than one dependent, or endogenous variables. This endogenes variable may appear in one equation as endogenous and as exogenous in another equation of tne system. Due to presence of the endogenous variables as explanatory variable in some of the equations it becomes correlated with the disturbance term of the equatioh in which it appears as the explanatory variable.