The error function is the

sum of squared errors between the network outputs and the target values, computed as in Equation (5).

0's) and then using Solver to achieve a minimum value of the

sum of squared error term (SSE; cell E2; highlighted in red).

In these months, a big part of the total

sum of squared errors occurs in the first four months of recessions.

The

sum of squared errors is minimized by equating the derivative of the squared errors to zero.

4) Convergence is monitored by observing

sum of squared errors of the observed values and corresponding predicted values from step (2).

M taken in the same [DELTA]t interval-minimizing the

sum of squared errors [between observed ([E.

MARS would take money growth, output growth, and the rate of currency depreciation as predictors and attempt to fit the best model for the inflation rate by placing knots and choosing additive and interactive effects to minimize the

sum of squared errors.

Learning is complete when no further reduction in the

sum of squared errors is possible.

xB[left arrow]C * X to create the B-coefficients, the predicted C values, CH, the errors, E, and the

sum of squared errors, SSE, in one step.