The extent to which a model or method is useful in making forecasts. This is best assessed by comparing it with alternative methods. Determining predictive validity has long been recognized as one of the primary ways to test hypotheses (Friedman 1953). See Armstrong (2001d).<o:p></o:p></P>
<ul>
  <li>Friedman, M. (1953), “ The methodology of positive 
economics,” in <I>Essays in Positive Economics.</I> Chicago: U. of Chicago 
Press.</li>
  <li>Armstrong, J. S. (2001d), “Evaluating forecasting 
methods,” in J. S. Armstrong (ed.), <I>Principles of Forecasting.</I> Norwell, 
MA: Kluwer Academic Press.</li>
</ul>