### Table 5 Kolmogorov-Smirnov Test to see if residual qualities of the returned units were consistent with the theoretical distributions.

"... In PAGE 16: ... Bars are experimental observations in Experiment 1, and curves are theoretical predictions, which are normalized to have the same masses as in the experiment. Furthermore, Kolmogorov-Smirnov Tests ( Table5 ) show that, in three out of four experiments, the distribution of residual qualities of the returned units is consistent with model predictions. Experimental evidence not only confirms the cherry picking phenomenon in a qualitative fashion, but also suggests that the theory is sound quantitatively despite all the handicapping factors mentioned before.... ..."

### Table 2 Comparison of causally ordered multi-channel algorithms Causal

### Table 5: Distribution of Causal Factors (dichotomous)

2004

### Table 1 Comparison of core/periphery fitness measures using Beck et al. (2003; ND) data

2004

"... In PAGE 5: ....P Boyd, W.J. Fitzgerald, R.J. Beck/Social Networks columns 4 and 5 of Table1 . Column 6 of Table 1 compares the results from the UCINET (Version 6.... In PAGE 5: ... For all 12 groups, all three of these algorithms matched the exhaustive search by consistently finding the global optimum from several starting configurations. [ Table1 about here] From the results in Table 1, the genetic algorithm in UCINET finds the global optimum in two out of our 12 cases. The UCINET fit statistic is among the five best for seven of the 12 cases, and among the ten best for nine of the 12 cases.... In PAGE 5: ... For all 12 groups, all three of these algorithms matched the exhaustive search by consistently finding the global optimum from several starting configurations. [Table 1 about here] From the results in Table1 , the genetic algorithm in UCINET finds the global optimum in two out of our 12 cases. The UCINET fit statistic is among the five best for seven of the 12 cases, and among the ten best for nine of the 12 cases.... In PAGE 7: ... A low probability along with an intuitively high observed fitness value suggests that the observed data may have a core/periphery structure. To illustrate this permutation test, we used Mathematica to program a random permutation generator based upon the observed within group distribution of messages for each of the 12 groups from Table1 . As with the observed data, diagonal cells were also ignored for these permutations.... In PAGE 7: ... For Group 1, for example, no random permutation in each of the 3 runs produced an optimal fitness value equal to or greater than the observed fitness value of 0.867 (see Table1 ). For Group 3, 43 of the random permutations in the first run produced optimal fitness values equal to or greater than the observed fitness value (0.... ..."

Cited by 1

### Table 4: The encoding of causal terms

"... In PAGE 22: ...e., CCS terms), Table4 gives us an encoding of the classical causal bisimulation of CCS [DD89, Kie91] into the observation equivalence of the monadic -calculus. It is straightforward to see that there is agreement between the de nitions of the encoding on processes and on sortings: Proposition 5.... ..."

### Table 5: Causal Case [1]

2007

"... In PAGE 6: ...able 4: Random Case [1].................................................................................................... 22 Table5 : Causal Case [1].... In PAGE 21: ... The other option is that if at any letter, none of the 30 possible steppings for the cipher rotors yields a consistent path, that path is eliminated since it is a random path. Table5 was generated using the same method. However, instead of a random setting, the causal setting was used.... In PAGE 21: ...5 paths and a maximum of 84 paths. For the causal case, Table5 shows that for a 30-letter message, we expect 29.6 paths with the 10,000 cases tests.... In PAGE 22: ...Table 5: Causal Case [1] From the information in Table5 , we could reduce the number of random settings by saving only those settings that meet some threshold. One example would be if the setting exceeds the expected mean in the causal case.... ..."

### Table 1. Model Parameters for Observed Electron Distribution

"... In PAGE 5: ... Generation of EMIC Waves By the Observed Electron Distribution [18] To begin our investigation, we fit the distribution function of Figure 2 using a nonlinear least-squares fitting routine. In Figure 4 we show the model fit consisting of a background and a drifting Maxwellian with fitting parameters shown in Table1 . The parameters shown in Table 1 are those obtained from the best fit, except that we have increased the beam drift velocity from 8.... In PAGE 5: ... This would justify increasing the beam velocity by at least a factor of two for growth at lower altitudes and f gt; 50 kHz (the local cyclotron frequency during the Polar observations). [19] The parameters of Table1 were used in the analysis of wave growth using the modified WHAMP code. We have investigated the growth of EMIC waves for a number of frequencies.... ..."

### Table 1: Mean square error for monotonic regression in two explanatory vari- ables using the generalised PAV algorithm (GPAV) and a simple averaging technique (SA). The table shows mean values for 100 data sets, each consist- ing of 400 observations.

"... In PAGE 5: ... The error terms epsilon1 were either normally or exponentially dis- tributed with variance 1. Table1 shows that, regardless of the distribution of the error terms or the correlation between the two explanatory variables, the generalised PAV algorithm performed better than the simple averaging tech- nique. Furthermore, the difference in goodness-of-fit was particularly large for heavy-tailed (exponentially distributed) error terms.... ..."

### Table 3. Granger causal relationships from lagged anomalies of (a) temperature and (b)

"... In PAGE 21: ...We use the Granger causality algorithm to test the above hypothesis1. The results generally indicate significant causal relationships from lagged anomalies of precipitation and temperature to current precipitation variability ( Table3 , for time lags of two months). The nature of these relationships is given by the signs of the regression coefficients associated with the mean precipitation/temperature anomalies over the lagged period.... In PAGE 21: ...1) in the other months, with the only exception in December (Figure 7b). Together, Table3 and Figure 7 indicate that in summer precipitation anomalies have a positive relationship with their previous variations, while the relationships between 1 Here by saying that lagged precipitation anomalies Granger causes current precipitation variability, we simply mean that the former contains information about the latter that is not provided by other lagged variables (e.... In PAGE 22: ... The oscillatory variations in vegetation, which are captured by the autocorrelation analysis (Figure 4), the composite plots (Figure 5), and the spectral analysis (Figure 6), further support the proposed interactions of vegetation with soil moisture. Finally, the causal relationship between temperature and precipitation ( Table3 ) is consistent with the implied role of soil ... ..."

### Table 1. Left panel: basic data statistics; right panel: recommendation algorithm performance 4.2 Results and Discussions The observed values and the simulated mean and standard deviation of C4 and C6 are reported in Table 2. We also report the z-score to indicate how much the observed values deviate from the simulated distribution. Our current simulation generates 20 graphs from the uniform graph distribution conditional on the actual degree distributions. The observed values of C4 of the retail and book datasets both deviate significantly from the simulated distributions. The z-scores indicate that the observed C4 of the retail dataset deviates much more significantly than that of the book dataset does (141.38 as compared to 23.35), which is consistent with the general superior performance of user/item-based algorithms compared to the top-N most popular recommendation for the retail dataset. Dataset Topological Measure Observed Simulated Mean Simulated Stdev Z-score

"... In PAGE 5: ...Table1 presents the basic data statistics and the performance of the four recommendation algorithms measured by commonly used metrics of precision, recall, F measure, and rank score [3] that measure the accuracy, coverage, and ranking quality by matching the recommendation lists (10 products for each consumer) with withheld 20% later actual purchase records. Consistent with our earlier discussion, we observe that no one algorithm outperformed others for both datasets.... ..."