(MSE). The sum of the squared forecast errors for each of the observations divided by the number of observations. It is an alternative to the <a href="search.pl?p=mean absolute deviation">mean absolute deviation</a>, except that more weight is placed on larger errors. (See also <a href="search.pl?p=Root Mean Square Error">Root Mean Square Error</a>.) While MSE is popular among statisticians, it is unreliable and difficult to interpret. Armstrong and Fildes (1995) found no empirical support for the use of the MSE or RMSE in forecasting. Fortunately, better measures are available as discussed in Armstrong (2001d).<o:p></o:p></P>
 <ul>
   <li>Armstrong, J. S. &amp; R. Fildes (1995), “On the 
selection of error measures for comparisons among forecasting methods,” 
<I>Journal of Forecasting</I>, 14, 67-71. (<a href="../../../ideas/pdf/selectionerror.pdf">Full 
text</a>)</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>