### Table 1 shows empirical powers for the test statistics

1986

Cited by 5

### Table 24: Author Collaboration Networks - Statistics Power Law Exponent Clustering Coefficient

2007

"... In PAGE 22: ...e considered a Small World Network. The Newman clustering coefficient approaches 0.5, thus it can be concluded that the network is almost a Small World network according to Newman. How does this compare to other research and other digital collections? The results of other research is included in comparison to our findings for the ACL Anthology Network in Table24 . Please note that the results from other research may not include matching algorithms used to find certain values.... ..."

### Table 1: Fitted Power Law Parameters

1997

"... In PAGE 4: ... We observe that the mean call probability to the rank r callee, Pr, can be modeled using a power or generalized Zipf apos;s law at all three reference time periods: Pr apos; A rp , where A is the scaling parameter and p is the exponent parameter. Table1 shows the tted parameters and mean square errors of the ts. Figure 3 and Figure 4 are linear and log-log plots of Pr versus callee rank for the three reference periods.... In PAGE 15: ...4% 46.3% count per sec Table1 0: Simulation Results and Percentage Di erence for Centralized Database Scheme... In PAGE 17: ...1 4,743.8 rate (per sec) Table1 1: Access Rate at Three Selected Simulation Times... In PAGE 19: ...ugust 1990. S.F. Bay area transportation measurements. [25] International air tra c between united states and other countries. National Transportation Statistics 1996: Table1 7,18. [26] C.... ..."

Cited by 48

### Table 3. Broken power{law ts with galactic and free absorption. For these datasets, the broken power-law model is statistically preferred with respect to a single power-law model (the F{tests in the last column are against the single power{law model with free absorption)

2001

"... In PAGE 4: ... 1989). The results from the single power-law ts are reported in Table 2, while Table3 reports the broken power-law ts parame- ters. Adopting a conservative approach, we assessed the presence of a spectral break when the broken power-law model was signi cantly better, according to the F{test, even with respect to the single power{law model with free absorption.... In PAGE 5: ...rrors at 68% conf. level for 3 and 4 parameters of interest ( 2 =3:52 and 4.72 respectively). adopting an absorption value greater than the galactic one. In most cases, however, the broken power{law models with galactic absorbing columns provided even better ts (according to the F{test, see Table3 ). For 1ES 0033+595 and PKS 0548{322, instead, there is evidence for a higher column density also using the broken power{law models.... In PAGE 5: ... The spectrum is characterized by a curved shape, which is better t- ted by a broken power-law than by a single power-law model (with both xed and free absorption). The best t, however, still requires a column density in excess of the galactic value (see Table3 ), with high signi cance (F{test gt; 99:8%). The spectral indices before and after the break Fig.... In PAGE 10: ....2.7. H 2356{309 The spectrum of this source is not compatible with a sin- gle power{law model: there is clear evidence for a convex spectral shape, that is best tted by a broken power{law with high signi cance ( gt;99.9%, see Table3 ). This locates the synchrotron peak around 1.... ..."

### Table3-1. Analog VLSI vs.Digital VLSI ComputingPerformanceComparison

1996

### Table 3: Comparison of object detection in the retina pictures using the basic approach and

"... In PAGE 4: ...able 2. However the GA approachachieved 100#25 precision and recall. 5.3 Retina Pictures The results for the retina pictures are summarized Table3 . Compared to the results for the other two image databases these results are disappointing.... ..."

### Table 1. Multi-resolution data (geometry and images) used in the Bamiyan project.

"... In PAGE 6: ... A great force for this trend has been the availability and improvement of image and range sensors, as well as the increasing power of computers for storage, computation and rendering of the digital data. The Bamiyan project is a combination of multi-resolution and multi-temporal photogrammet- ric data, as summarized in Table1 . The geometric resolution of the recovered 3D data spans from 20 m (SPOT5) to 5 cm (Buddha model) while the texture information is between 2.... ..."

Cited by 1

### Table 2: Regression estimates of power law log y = log a + b log x trading

2001

"... In PAGE 14: ...As indicated in Table2 , the parameters of the distribution were significantly different at each of the time steps for which that data is represented. The differences are statistically significant at any confidence level up to five significant digits.... ..."

Cited by 2

### Table 3 Parameters of Best-Fit Functions for Digit-Entering Task Empirical and Simulation Results

in Theoretical and computational analysis of skill learning, repetition priming, and procedural memory

2002

"... In PAGE 6: ... This follows from the definition of skill learning as being the improvement in performance on unique stimuli over blocks. What can we say about performance on repeating stimuli? Data pertinent to this question come, once again, from Kirsner and Speelman (1996), who showed that power functions also provided an excellent fit to performance on the repeating words and non- words in their lexical decision task (Kirsner amp; Speelman, 1996, Figure 5, and Table3 ). Further evidence that multiple repetitions give rise to a reaction time function that follows the power law comes from Logan (1990, Table 1, and Figures 5 and 6).... In PAGE 25: ..., 1999, Experiment 1). Parameters of the best-fit power functions are shown in the upper half of Table3 , along with two goodness-of-fit measures, root-mean-squared deviation (RMSD) and r2, which show that these power functions provide a good fit to performance on unique stimuli and performance on repeating stimuli in the digit-entering task. For IRP, we found that a linear power function provided as good a fit in terms of RMSD and r2 as did the best fitting nonlinear power function.... In PAGE 25: ... For IRP, we found that a linear power function provided as good a fit in terms of RMSD and r2 as did the best fitting nonlinear power function.8 This best linear fit is also graphed in Figure 13a, and its parameters are shown in the upper half of Table3 . These fits to empirical data from the digit-entering task exhibit the same pattern as that obtained by Kirsner and Speelman (1996): Perfor- mance on unique stimuli and performance on repeating stimuli follow the nonlinear power law, but repetition priming is linear.... In PAGE 25: ... For IRP, however (again, as with the empirical data), a linear power function provided as good a fit as did the best fitting nonlinear power function. The parameters of these fitted functions are shown in the lower half of Table3 , along with the goodness-of-fit measures. Even though these measures arise from a single learning mechanism in the model, they exhibit the same pattern of dissociation as in the digit-entering task empirical re- sults, viz.... In PAGE 26: ...974 and .950, respectively ( Table3 ). Similarly, the RMSD and r2 values for the comparison between empirical and simulated performance on repeating stimuli curves, Figure 13a vs.... ..."

Cited by 5

### Table 3. Descriptive Statistics for the Grade 2 Empirical Data and 12 Sets of Simulated Data Growth Amount Mean SD Max Min Skewness Kurtosis

2004

"... In PAGE 13: ... Results Simulated theta distributions. Table3 contains the descriptive statistics for the 12 type x amount growth conditions. Vertically Articulated Standards 13 ... In PAGE 14: ...The means and standard deviations in Table3 are as expected: Negative growth means are lower than no growth means, low and moderate growth means are higher than no growth means, standard deviations for the grade 2 empirical data and the linear and remediation growth means remain unchanged, the standard deviations for the remediation growth model are slightly reduced by the reduction in skewness and shift to the right introduced by the growth model, and the standard deviation for the Rich Get Richer growth model increase somewhat, also by the reduction in skewness and shift to the right introduced by the growth model. The grade 2 empirical mean and standard deviation reflect the effects of re-centering of the full grade 2 item bank on the items in the ordered item booklet used in standard setting.... ..."