### Table 5. Performance of the Varimax and Partial Least Squares Estimators

"... In PAGE 21: ... The means of the 100 squared distances (empirical MSE estimates) for each of the other PCR methods and each of the 32 simulation settings are provided in Table 4. Analogous results for the varimax and PLSR estimators are provided in Table5 . For each of the 32 simulation settings, the OLSR estimator along with the other estimators discussed in Section 6 were ranked according to their empirical MSE values with lowest ranks corresponding to lowest empirical MSE.... ..."

### Table 3. Principal, Varimax, and Partial Least Squares Components

"... In PAGE 20: ... Methods 3(c) and 3(d) use the first two PLS components to estimate a250 . The coefficients of the linear combinations that provide these components are found in the last two lines of Table3 . These PLS compo- nents are more difficult to interpret than the principal components or the rotated principal components in this example.... ..."

### Table 3. Effects of Full reform (vis-a-vis partial or no reform) on mainline penetration and productivity

2002

"... In PAGE 18: ...olicies are in place and zero otherwise (i.e., the three way interaction term).36 We find this variable to be highly significant for both mainline penetration and productivity ( Table3 ). The estimated coefficients suggest that mainline penetration is 8 percent higher and productivity is 21 percent higher in years of complete reform compared to years of no or partial reform.... ..."

Cited by 4

### Table 3. Summary of ANOVAs with Factors Condition and Time for the First and Second Level Analysis of the Theta Effecta

"... In PAGE 7: ...vs. second) calculated for the mid-frontal ROI (see Table3 for F and p values as well as for effect size measures partial eta squared Z2 p and omega squared o2). Furthermore, theta power increased significantly toward the end of the musical pieces of both categories as statistically confirmed by a significant main effect of Time.... In PAGE 7: ... In this second level analysis, theta power increased only during the second half of pleasant pieces compared to base- line and remained unchanged in the course of unpleasant ex- cerpts. This was statistically supported by a significant two-way interaction (see Table3 ) in an ANOVA with factors Condition (pleasant vs. unpleasant) and Time (first vs.... ..."

### Table 7. Effect of group size on the percentage of members who assist to assemblies (Partial results from robust ordinary least squares regression)

"... In PAGE 7: ... We can explore further the role of the size of the community in affecting the quality of cooperation in 1997 by introducing a spline in the relation between assistance to assemblies and the number of ejidatarios.3 Seeking by tatonnement the optimum breaking points in the number of ejidatarios on a best fit basis, we locate those at community sizes of 50 and 75 ( Table7 ). Results are represented in Figure 2.... ..."

### Table 1. Summary of ANOVAs. ANOVAs were conducted with four regions of interest on the logarithmic FFT data with factors stimulation (stimulus, rest), type (music, EEG question- naire), anterior-posterior, hemisphere, and group (without mIED, with mIED).

"... In PAGE 5: ... Moreover, a marginally significant interaction between factors stimulus and group was found (p lt; .07), reflecting that the de- crease of theta power during stimulus condi- tions was stronger in the mIED group (see Table1 for results of ANOVAs, see Methods for explanation of ANOVA model). When tested one-sided (justified by the hypothesis of a decrease of theta oscillations, see Introduc- tion), ... ..."

### Table 6: Sum of Squared Difierences Between the Estimates when Dividing the HIV Data in Two Groups of 4 Replicates. The posterior mean reduces the variability by 83% relative to the raw log ratios.

2003

Cited by 2

### Table 6: Comparing modified least squares (dual) and SVM on dataset msweb. For expla- nation of entries refer to Table 2.

2002

"... In PAGE 15: ...5, the resulting partial class separation is sufficient to yield useful ranking information when we compare different items. Table6 compares the performance of the SVM and the modified least squares (dual) formulations on the msweb dataset.... ..."

Cited by 10

### Table 12. Percent variation accounted by Partial Least Square factors for the Coastal Plain (CP) ecoregion. This model is not significant and therefore it did not have the sample validation for the number of extracted factors.

2002

"... In PAGE 46: ...325 c) Coastal Plain (CP): The scarcity of sampling sites (n=7) in the Coastal Plain precludes an analysis like those completed for the Blue Ridge and Piedmont areas. Therefore, we only presented the percent variation that accounted for two PLS factors in Table12 . X_ and Y_Scores and weights for the model are shown in Figures 13 amp; 14.... In PAGE 46: ... Percent forest, percent crop, and percent pasture were the least important (VIP lt; 0.8; Table12 ) landscape metrics in the Coastal Plain. Soils in this ecoregion are generally of low erodibility, and the terrain is much flatter than the other two ecoregions, hence, area on slope gt; 3% was not significant as in the Blue Ridge and Piedmont ecoregions.... ..."

### Table 8. The significant factors of the final PLS model for the 5 biota and 14 landscape metrics and percent variation accounted by Partial Least Square factors for the Blue Ridge (BR) ecoregion.

2002

"... In PAGE 42: ...By Ecoregion The final model with the 14 independent variables was rerun by ecoregion. a) Blue Ridge (BR): There was only one significant factor that accounted for 17% and 74% of the variability for the biota and landscape metrics, respectively ( Table8 ). Ag_mod, Ag_slp, Ag_slp_mod, Past_slp, percent forest, percent pasture, total area with slope gt;3%, Sd_slp, Mean_slp (VIP gt;1) were the most important variables followed by the percent crop and stream density (0.... ..."