It is generally held that by means of statistical and mathematical methods one can organize historical data into a useful body of information, which in turn can serve as the basis for the assessments of the state of the economy. It is also held that the knowledge secured from the assessment of the data is likely to be of a tentative nature since it is not possible to know the true nature of the facts of reality.
Some economists such as Milton Friedman held that since it is not possible to establish “how things really work,” then it does not really matter what the underlying assumptions of a theory are. On this way of thinking, what matters is that the theory can yield good predictions.
According to Friedman,
The ultimate goal of a positive science is the development of a theory or hypothesis that yields valid and meaningful (i.e., not truistic) predictions about phenomena not yet observed…. The relevant question to ask about the assumptions of a theory is not whether they are descriptively realistic, for they never are, but whether they are sufficiently good approximation for the purpose in hand. And this question can be answered only by seeing whether the theory works, which means whether it yields sufficiently accurate predictions.1
For instance, an economist forms a view that consumer outlays on goods and services are determined by disposable income. Based on this view he forms a model, which is then validated by means of statistical methods. The model is then employed in the assessments of the future direction of consumer spending.
If the model fails to produce accurate forecasts, it is either replaced, or modified by adding some other explanatory variables. What matters here is how well consumer outlays are correlated with various variables in order to secure a good predictive model. In this sense, all that an economist requires is to establish a good fit between dependent variable and various other variables.
On this way of thinking, we form our view regarding the real world based on how well the various pieces of information are correlated with each other.