Research Article

Robust Wavelet Estimation to Eliminate Simultaneously the Effects of Boundary Problems, Outliers, and Correlated Noise

Table 4

Simulation results of global and local mean squared using AR with different parameters. The first half of the table displays the global mean squared error.

ParameterI II III

CD
1.16741191.14304050.6588707
1.17539151.14562390.8297044
1.42716011.50234871.0724085
0.95751550.95640910.8454300

SQ
1.7787951.10638261.570100
1.6936341.04498331.736956
1.8825621.26807051.862326
1.5960860.94449051.669918

AQ
1.8984001.2352521.742275
1.7712811.1291601.694660
1.9049991.2890101.979608
1.8601641.2972061.929902

Fat tail data
1.11547690.99816090.4987108
0.94029730.88669340.6799741
1.17248181.26016061.1130583
0.9557715 0.92382370.8206673

mse at boundary

CD
3.31961293.31669441.1779039
1.07570301.13000480.6423341
1.03650452.03575430.9276551
0.63781570.74005420.6481636

SQ
3.14392082.51694182.102952
0.96897820.38687471.548121
1.82438691.06613101.839487
1.81547541.24519242.386717

AQ
1.5765870.99098151.885401
2.2588081.13071862.460695
1.8725681.09073022.251098
1.5323321.25938572.012984

Fat tail data
3.12545642.58213310.5579562
0.72358220.74571320.7487358
0.62877450.74757770.7258418
1.26957451.20632350.9366066