Research Article

Coercively Adjusted Auto Regression Model for Forecasting in Epilepsy EEG

Table 5

Comparison of the foresting error rates and the execution time between the existing methods and the proposed method.

Forecasting
time points
ElectrodesRMSEExecution time (sec)
500 (training)1000 (training)500 (training)1000 (training)
ANNNNCA-AR (5)ANNNNCA-AR (5)ANNNNCA-AR (5)ANNNNCA-AR (5)

100040.0910.0200.0470.130.0230.051.017.440.085.349.110.06
80.0390.0060.0430.0550.0160.0510.087.750.035.018.860.06
350.2280.0720.0390.1920.0620.0450.907.640.084.908.780.05
700.0570.0100.0560.0490.0110.0540.107.570.084.488.630.06
950.0910.0320.0440.1450.0380.0340.947.440.064.708.750.05

Average0.1010.0280.0460.1140.030.0470.617.570.074.888.830.06

200040.2170.0210.0460.2450.030.0381.1120.550.084.8720.140.05
80.0600.0190.0470.0820.0270.0181.1120.640.085.5520.870.05
350.2530.0630.0520.1910.0880.0480.8320.310.064.9121.030.05
700.0580.0150.0490.0460.0120.0331.0120.120.084.5220.940.04
950.1730.0370.0480.1980.0380.0480.9220.580.085.1220.470.05

Average0.1520.0310.0480.1520.0390.0371.0020.440.075.0020.690.05

300040.3760.0290.0520.3990.0190.0510.9232.740.085.7938.310.05
80.0940.0150.0510.1120.0320.0511.2232.390.035.6836.470.05
350.2590.0700.0490.2130.0870.0470.9832.400.084.7636.570.05
700.0550.0230.0480.0440.0130.0461.3332.790.085.1337.530.05
950.2270.0490.0500.2330.0590.0490.9433.290.114.8837.470.05

Average0.2020.0370.0500.20.0420.0491.0832.720.075.2537.270.05