Econometrics

Econometrics is the area of statistics specialized to deal with economic models.  We can identify distinctions between both economics and econometrics and between statistics and econometrics.

Economics (as typified by the Classic Economic Models featured on this web site) is generally presented in terms of smooth curves and precise equations.  Reality is a cloud of data points that may only suggest the existence of some underlying smooth curve.  The reconciliation of these two views takes the form of a model

y = a + b x + e,

where x and y are observable data, a and b are parameters, and e is an error term.  The relation

y = a + b x

would be a smooth line.  The error term e accounts for why the observed data points do not lie on that straight line.  The goal of econometrics is to make inferences about a and b given the observed cloud of data points (x,y).

The major distinction between econometrics and statistics is that economic models almost universally involve multiple equations.  As a consequence, the typical economic model involves several equations such as

y1 = a + b y2 + c x + e,

where both y1 and y2 are endogenous (or determined by the model at hand.  The range of statistical models, on the hand, includes many models where right-hand side endogeneity is not an issue.  Econometrics also involves equations such as

y1 = a + b y2* + c x + e,

where y2* is the expectation of y2 rather than its observed value.  This kind of modeling elements takes econometrics into areas not at all common in statistical models of, for example, a science experiment where the explanatory variables are neither endogenous or determined by the expectations of the subject of the experiment.

Classic Economic Models
Interactive presentations of the most important models
in microeconomics and macroeconomics go beyond
anything appearing in a printed-on-paper textbook.
Learn to think like an economist.