### Table 11: Observed number of non-linear iterations CXD8D2D0, grid point errors and respective orders of con- vergence for both steps of QSC on Problem 6.

2004

"... In PAGE 25: ...) Due to the simple shape of the solution, the application of the adaptive technique of Section 3 to Problem 6 resulted in a uniform partition, that is, the drift corresponding to the initial uniform partition was below the adaptive tech- nique tolerance D8D3D0 BP BDBCA0BE. Table11 shows that the QSC method applied to Problem 6 exhibits optimal convergence. It is worth noting that the second step of QSC requires fewer non-linear iterations than the first step, a fact which can be attributed to the choice of initial guess for the second step, as discussed above.... ..."

### Table 1. The Calculated Results for Analyzed Data-Set

2000

"... In PAGE 9: ... In order to have easy interpretable models, we have fixed the maximal number of terms in the equation to be equal to 8 and the maximum degree of polynoms to be equal to 3. The calculations performed using the select params option of the ANALYSIS are summarized in Table1 . The number of stored models was 3.... In PAGE 9: ... It was shown that the use of significant variables, as detected by MUSEUM, = improved PLS results (compare data in column 7 vs. column 6 in Table1 ). The similar tendency was also observed if only variables found to be relevant by the PNN algorithm were used in the cross-validation calculations (compare the last and 7 columns of Table 1).... In PAGE 12: ... b Number of significant PLS components. c The cross-validated q2 calculated using input variables optimized by MUSEUM approach (unless not stated otherwise the PLS results are from Table1 and 15 of (2)). d Number of input variables selected by PNN.... ..."

Cited by 2

### Table 1. Empirical Results

"... In PAGE 11: ...INSERT FIGURE 5 (A) TO (F) Empirical Results We present the results of the estimation in Table1 , together with the results of a linear regression of the selected macroeconomic variables on the devaluation probability. The linear regression might also be seen as the estimation of the model presented earlier with Eq.... In PAGE 11: ... The expected signs are, therefore, opposite to those assigned in the case of Jeanne apos;s model. Table1 shows that, for the non-linear case, the level of international reserves is the only variable that is significant and has the expected sign. This points to the importance of this variable in the determination of the fundamental of the Brazilian economy.... In PAGE 12: ...he evolution of the estimated fundamental can be seen in Fig.7. Fig.9 shows the separate contribution of each macroeconomic variable in the composition of the fundamental. One can see clearly the importance of the level of international reserves in this composition, in accordance with the empirical results presented in Table1 . Observing Fig.... ..."

### Table 1 Comparison of the non-linearity between the optimized and the random con quot;gurations

"... In PAGE 13: ... In any case, the RMS of the non-linearity distribution is K1% at most. Using tower-dependent preshower weight fac- tors, we compare the non-linearities between the two con quot;gurations in Table1 . For the random con quot;guration, we calculate the average non-lin- earity and the RMS for each con quot;guration out of 100 trails.... ..."

### Table 6 Optimal trading strategy under non-linear permanent market impact with lag

"... In PAGE 27: ...1. Solution in Table6 shows less pronounced differences with the naive strategy due to the convexity of the non-linear impact functions, which implies smaller execution costs in comparison to the linear non-convex permanent market impact. 4 Conclusions In this paper, we have developed an approach for optimal transaction execution based on the stochastic programming framework and a sample-path scenario model.... ..."

### Table 6: Optimal trading strategy under non-linear permanent market impact. c =1.0

2003

"... In PAGE 27: ... Table6 shows less pronounced differences with the naive strategy due to the convexity of the non- linear impact functions, which implies smaller execution costs compared to the linear non-convex permanent market impact.... ..."

Cited by 2

### Table 3. Model Fit Errors and Prediction Errors for Non-Linear Process

"... In PAGE 26: ... The fit of all of the models to the data is shown for a 140 point segment in Figure 23. The fit errors (sum of squared residuals) are given in Table3 . Note that the fit error of all the non-linear models and transformed linear model are about equal.... In PAGE 27: ...26 each model. The PRESS numbers are given in Table3 . In this test the FIR model identified using non-linear PLS model was found to have the smallest prediction error, followed by the non-linear PCR model and the polynomial regression.... ..."

### Table 2: Non-linear Data sets with f(x)=x

"... In PAGE 17: ... In these data sets the true R 2 was chosen to be larger than in the linear data sets in order to have a comparison in this range. But the multiple regression method was given the strong advantage of having the correct nonlinear function (see Table2 ), so only the coe cients and the constant term of the three non-linear components of the regression had to be estimated. With this considerable advantage the linear method is surely an optimal or nearly optimal... ..."

### Table 1: Throughput optimization of non-linear real-life designs with no unfolding: ICPL - initial critical path length

"... In PAGE 4: ... 6 Experimental Results We have used several Volterra filters of different orders, several Noise Sharpers, Echo Canceller, Vocoder, NEC DAC (digital-to- analog converter), and ADPCM (adaptive delta pulse code mod- ulation) for experimental results. Table1 shows throughput im- provement achieved using the best previous approach by [2] and using the new approach with optimization engine of general non- linear computation with no unfolding. Table 2 shows throughput improvement achieved for non-restricted amount of unfolding.... ..."

### Table 1: Throughput optimization of non-linear real-life designs with no unfolding: ICPL - initial critical path length

"... In PAGE 4: ... 6 Experimental Results We have used several Volterra filters of different orders, several Noise Sharpers, Echo Canceller, Vocoder, NEC DAC (digital-to- analog converter), and ADPCM (adaptive delta pulse code mod- ulation) for experimental results. Table1 shows throughput im- provement achieved using the best previous approach by [2] and using the new approach with optimization engine of general non- linear computation with no unfolding. Table 2 shows throughput improvement achieved for non-restricted amount of unfolding.... ..."