### Table 3: Stable Coalition Structures*

"... In PAGE 14: ...2.3 Results Table3 lists internally and externally stable coalition structures that are stable under at least one scenario. X means that this coalition structure is internally and externally stable (as defined in section 4.... In PAGE 14: ...3. Table3 about here Whereas in the no transfer case no coalition was stable, now at least one coalition is stable under every transfer scheme. With exception of scenario 4c, only one or two coalitions are stable.... In PAGE 14: ... Nevertheless, also small coalitions can make a difference by closing the gap between no and full cooperation under most scenarios by 50 or more percent. Under most scenarios, the number of violations VISi as listed in Table 2 already provides a good indication which regions participate in a stable coalition as listed in Table3 : only regions with a low VISi number form stable coalitions. It is interesting to observe that no coalition including only the key industrialized regions USA, JPN and EU is stable.... In PAGE 14: ... This indi- cates that it is not that straightforward to establish stable cooperation between industrialized countries, countries in transition and developing countries through transfers. As conjectured 8 Total transfers in the last column of Table3 are the sum of all positive transfers (=sum of all negative transfers) but not the sum of all transfers that is zero by definition. Total transfers are an indicator of the amount of financial resources redistributed by the transfer scheme.... In PAGE 16: ...3.2 Results In Table3 , X indicates stable coalition structures under exclusive membership where we restrict attention to unanimity voting. In the case of no transfers (scenario 0), the impact is dramatic.... ..."

### Table 1 Summary of Regression Analysis: Mean Reaction Times and Percentages of Errors by Condition Positive

1989

"... In PAGE 3: ... In all conditions, search time increased roughly linearly with display size. Table1 summarizes the linear regression data, and shows that the least squares fit to the average RTs accounts for over 99% of the variance in all conditions. Table 1 also shows the error rates for the positive and negative conditions.... In PAGE 7: ... In the simulation illustrated in Figure 5, thresholds were varied in each display composition so that the simulation produced error rates ap- proximating the actual error rates we observed. A comparison of the error rates given in Table 3, those generated by the simulation, with the error rates given in Table1 shows that the simulation captures the increase in false negatives with display size, as well as the relatively low and stable false- positive error rates. There is a tendency for the model to make more false-positive errors on larger display sizes, but this tendency is minimized by the ceiling effect of the almost perfect performance in the negatives.... ..."

Cited by 4

### Table 5: Results on stable Z-structures

2007

"... In PAGE 11: ... PERM was run in both directions, and in the case where neither direction finds the (known) lowest energy, the expected run-time is reported for the best energy found. The mean run-times for the Z-structures is reported in Table5 . REMC finds the unique conformation of each sequence relatively easily with a worst case mean run-time of 2 CPU seconds.... ..."

### Table 1: Weakly stable solution of Toeplitz systems

1993

"... In PAGE 11: ...0B is a positive regularisation parameter. Only small changes in equations #2820#7B23#29 are required. 7 Numerical Results The algorithm described in xx5#7B6 has been implemented in Pascal on an IBM PC and DEC VAX. In Table1 we give some results for randomly chosen n #02 n Toeplitz systems on a DEC... In PAGE 12: ... A conse- quence is that kxk=kbk is unlikely to be large, even if A is poorly conditioned, but this is typical of most applications. Table1 gives the condition number #14 1 #28R#29=kRk 1 #01kR ,1 k 1 ; which is a rough approximation to #14 2 #28R#29=#14 2 #28A#29 #28the 1-norm was used for computational convenience#29. Table 1 gives e 1 = k ~ R T ~ R , A T Ak 1 quot;kA T Ak 1 ; e 2 = k~ x , xk 2 quot;#14 1 #28R#29 2 kxk 2 ; e 3 = krk 2 quot;#14 1 #28R#29kAk 1 kxk 2 ; where r = A~ x , b.... In PAGE 12: ... Table 1 gives the condition number #14 1 #28R#29=kRk 1 #01kR ,1 k 1 ; which is a rough approximation to #14 2 #28R#29=#14 2 #28A#29 #28the 1-norm was used for computational convenience#29. Table1 gives e 1 = k ~ R T ~ R , A T Ak 1 quot;kA T Ak 1 ; e 2 = k~ x , xk 2 quot;#14 1 #28R#29 2 kxk 2 ; e 3 = krk 2 quot;#14 1 #28R#29kAk 1 kxk 2 ; where r = A~ x , b.From Theorem 1 and the bounds #2840#7B41#29, we expect these quantities to be bounded by #28low degree#29 polynomials in n.... In PAGE 12: ... The results con#0Crm this. For comparison, the last column of Table1 gives e c 3 , the value of the normalised residual e 3 obtained via Cholesky factorization of A T A. It can be seen that e 3 is not much larger than e c 3 .... ..."

### Table 3: The Di erences in Research Groups Stability Assessments by Managers in Un- stable and Stable Institutes (%) Types of Research Groups

1994

"... In PAGE 22: ... Among the surveyed people there were 30% working in two unstable institutes, and 70% in three relatively stable ones. But what is the actual di erence between military institutes of two types? Firstly, the unstable organizations are characterized by a high degree of instability in the personnel situation ( Table3 ).5 The di erences in assessments of stability of the research groups between the types of institutes are statistically signi cant: the share of stable groups in stable institutes is more than twice that of the unstable ones.... ..."

### Table 2: Relative stability scores obtained using MoJo. The greater the score, the more relatively stable the algorithm.

"... In PAGE 5: ... When comparing algorithms with regard to a con- catenation of different systems, we actually concatenated the similarity series from all those systems for each cluster- ing algorithm and then we applied Equation 3. The scores are given in Table2 . Those obtained using concatenation are listed in the ALL column and they are used to determine the overall stability ordering.... In PAGE 5: ... Those obtained using concatenation are listed in the ALL column and they are used to determine the overall stability ordering. All the scores in Table2 seem to tell the same story. In the direction of increasing stability, the chosen algorithms can be ordered as: Bunch, CL90, CL75, ACDC, SL75 and SL90.... ..."

### Table 6 The spatial distribution of first saccades from Experiment 2a

1998

"... In PAGE 7: ... This suggests that subjects maintained a relatively fixed and constant saccade latency time in this task, so task difficulty here had an impact primarily on first saccade accuracy. Table6 and Fig. 5 show the distribution of error saccades in this experiment.... ..."

### Table 7 Recommended contractor pay factors for relative compaction (Percentage of future rehabilitation cost in current-year dollars) As-measured standard deviation of relative compaction (%) As-measured average

1997

"... In PAGE 8: ... The following considerations dictate the specific recommendations herein: 1) One pay-factor schedule should apply to all new construction, that is, job-specific pay factors are undesirable; 2) The contractor should generally be charged a penalty for inferior construction which is out-of-specification, the magnitude of which should equal the full added cost to the agency of failure to meet the construction target; 3) The contractor should generally be awarded a bonus for superior construction which is within specification, the magnitude of which should be some fraction of the full added benefit to the agency resulting from the improved pavement performance (one half has been chosen herein as an appropriate point from which to start); 4) Pay-factor schedules should incorporate average and standard-deviation categories consistent with the accuracy within which estimates are determined from field measurements; 5) Because pay-factor schedules should be as simple as possible, increments of 5 percent in the bonuses/penalties seem appropriate; and 6) The standard deviations of pay-factor schedules must reflect expected testing and sampling errors as well as materials/construction variabilities. Recommended contractor pay-factor schedules, based on these guidelines, are presented in Table7 for relative compaction, Table 8 for asphalt content, and Table 9 for asphalt-concrete thickness. In developing these recommendations, average agency costs for the four pavement sections that were evaluated herein are assumed to be reasonably representative of typical new construction in California.... ..."

### Table 1 shows that the average equity-to-assets ratio is relatively stable in both IPO time

2002

"... In PAGE 7: ... We also exclude individual firm-year outliers for capital structure and the market-to-book ratio, as discussed below. Table1 reports summary statistics of capital structure, investment decisions (measured by the change in assets), and financing decisions. The variables are summarized in IPO time cross- sections, which hold the number of years since the initial public offering constant, and also in calendar time cross-sections.... In PAGE 11: ...ook (1.42 from Table 2) increases net equity issues by 4.97 percentage points, which is 33 percent of one standard deviation (15.05 from Table1 ) and 72 percent of the mean (6.86 in Table 1) in net equity issuance for firms of this vintage.... In PAGE 43: ...Table1 . Summary statistics of capital structure, investment decisions, and financing decisions.... ..."

Cited by 15

### Table 5: Types of R amp;D which were being Curtailed in Relatively Stable and Unstable Institutes (%)

1994