### Table 7 Sample ACF for Residuals and Squared Residuals

"... In PAGE 12: ...0 is first fitted. Table7 (upper panel) presents the sample autocorrelation function (ACF) of the residuals. It does not show any model inadequacy.... ..."

### Table 2 Residuals Statistics Estimation Residuals Statistic

### Table 5. Residual Correlations

1999

"... In PAGE 9: ... We use the original return data to calculate the regression residuals and find the residual correlations among the stocks. The residual correlation is reported in Table5 . We find relatively large residual correlation between eABX and eAEM and between eBGO and eAEM.... ..."

Cited by 3

### Table 5. Residual Correlations

"... In PAGE 14: ... We use the original return data to calculate the regression residuals and find the residual correlations among the stocks. The residual correlation is reported in Table5 . We find relatively large residual correlation between eABX and eAEM and between eBGO and eAEM.... ..."

### Table 11 Residual Diagnostics

"... In PAGE 28: ... (24) persistent movement of log volatility with high-frequency noise superimposed. The residual diagnostics in Table11 and Figure 4 indicate that the two-factor models are adequate. The measurement equation residuals are serially uncorrelated and virtually Gaussian.... ..."

### Table 11 Residual Diagnostics

"... In PAGE 26: ... (24) persistent movement of log volatility with high-frequency noise superimposed. The residual diagnostics in Table11 and Figure 4 indicate that the two-factor models are adequate. The measurement equation residuals are serially uncorrelated and virtually Gaussian.... ..."

### Table 3. Residual error

"... In PAGE 14: ... A vector which minimises the residue vector is accepted as a solution of the problem. The residuals of test examples 1 and 2 are listed in Table3 , for various n and m. The notation is the same as in Table 2.... In PAGE 14: ... Further iterations would not improve the result. It is discernible from Table3 that the minimum residuals are roughly O(1=n3=2) O(1=n2). The con- vergence speed and the solution accuracy are signi cantly higher if the boundary is smoother.... ..."

### Table 3: Residue Variances

"... In PAGE 13: ... The error residuals are computed as [9] ir(n) = ix(n) + NINT f M Xi=1 ?f;i(n ? 1) i?1(n ? 1)g (41) The RLSL usually will perform better than the GAL or the NLMS algorithms. As an example, the seismic waveforms given in Figures 5-6 were each subjected to all three methods; the error variances are given in Table3 . As can be seen, the RLSL outperformed the NLMS and GAL algorithms in nearly every case.... ..."