An exogenous variable is a factor that is outside of an economic model; it has an impact on the outcome of the model, but changes in the model do not affect these factors. Put simply, an exogenous variable is something that affects a particular outcome without being controlled by that outcome in return. Exogenous variables are sometimes referred to as independent variables, as opposed to dependent — or endogenous — variables, which are explained by the mathematical relationships in the model. While endogenous variables can be manipulated within the economic model, exogenous variables are generally uncontrollable.

One way to better understand exogenous variables is to look at a basic economic model, such as that for the supply and demand for a certain good in terms of the quantity demanded and supplied of that good for different price levels. In the basic version of this model, changes in the amount of money the consumer has to spend on the product affects the amount of demand for the good, but the demand curve does not influence the consumer's income. In this particular instance, consumer income is an exogenous variable.

The home computer is a consumer product that can be used as an example. There is a certain level of demand for home computers; manufacturers are able to supply the product at a certain price to meet this demand. The dependent variables in this example include the price of the computers and the number that are produced. Consumer income is an exogenous variable; how much money the consumer has to spend on the computer is not a part of the equation, but can affect the graph in that it alters the supply and demand curves — and therefore shifts the equilibrium level of price and quantity. It is important to note that a particular variable may be endogenous to one model but exogenous to another.

Another example of an exogenous variable is measurement error. Measurement error is the difference between the actual value of a quantity and the value measured experimentally. Some measurement error is generally taken to be a given in most models.

Path Analysis, the study of causation, causality, and correlation between different variables, investigates the interrelatedness of exogenous variables, and the ensuing effects on endogenous variables. These influences can either be direct or indirect, and are portrayed through structural equation modeling. Structural equation models are constructed by summing the contributions of all exogenous variables to a specific outcome.