### Table 1: Loss probabilities for di erent values of b under both synchronized and uniformly distributed frame boundaries.

1995

"... In PAGE 21: ... These results illustrate the importance of incorporating time- scales into tra c models in order to accurately assess the impact of the modeled tra c sources on network performance. In order to investigate a more realistic multiplexing scenario, Table1 depicts cell loss prob- abilities for a bu er capacity of B = 50, an output link capacity of 100 Mbps, and an input consisting of 10 identical VBR sources each transmitting at a rate of 8 Mbps. This scenario can be modeled with GPM sources through the following choice of parameters: Ts = 327,... In PAGE 22: ...Table 1: Loss probabilities for di erent values of b under both synchronized and uniformly distributed frame boundaries. In Table1 , three di erent autocorrelation functions for cells/slice are considered. One set of results represents the frame-level pre-bu ering case (b = 0) while the others correspond to two of the autocorrelation curves pictured in Fig.... In PAGE 23: ... The impact of frame-boundary synchronization can be minimized by uniformly distributing the frame starts (boundaries) at intervals of bMN c, as is the case in [11], which represents a best- case scenario in terms of source starting times. While ensuring such a uniform distribution of frame starts in practice is highly unlikely, the positive impact on the multiplexer is indisputable, as is seen in Table1 , where loss probabilities virtually coincide for all values of b. Recall that the results presented here are for one possible combination of frame cell rates, and that under a quasi-static approximation, all combinations of the individual frame cell rates would be considered.... ..."

Cited by 3

### Table 1: Loss probabilities for di erent values of b under both synchronized and uniformly distributed frame boundaries.

1995

"... In PAGE 21: ... These results illustrate the importance of incorporating time- scales into tra c models in order to accurately assess the impact of the modeled tra c sources on network performance. In order to investigate a more realistic multiplexing scenario, Table1 depicts cell loss prob- abilities for a bu er capacity of B = 50, an output link capacity of 100 Mbps, and an input consisting of 10 identical VBR sources each transmitting at a rate of 8 Mbps. This scenario can be modeled with GPM sources through the following choice of parameters: Ts = 327,... In PAGE 22: ...Table 1: Loss probabilities for di erent values of b under both synchronized and uniformly distributed frame boundaries. In Table1 , three di erent autocorrelation functions for cells/slice are considered. One set of results represents the frame-level pre-bu ering case (b = 0) while the others correspond to two of the autocorrelation curves pictured in Fig.... In PAGE 23: ... The impact of frame-boundary synchronization can be minimized by uniformly distributing the frame starts (boundaries) at intervals of bMN c, as is the case in [11], which represents a best- case scenario in terms of source starting times. While ensuring such a uniform distribution of frame starts in practice is highly unlikely, the positive impact on the multiplexer is indisputable, as is seen in Table1 , where loss probabilities virtually coincide for all values of b. Recall that the results presented here are for one possible combination of frame cell rates, and that under a quasi-static approximation, all combinations of the individual frame cell rates would be considered.... ..."

Cited by 3

### Table 1: Task is to nd a probability distribution p under constraints p(x; 0) + p(x; 1) = :6, and p(x; 0) + p(x; 1) + p(y; 0) + p(y; 1) = 1

1997

"... In PAGE 4: ... (The constraint that Pa;b p(a; b) = 1 is implicit since p is a probability distribution.) Table1 represents p(a; b) as 4 cells labelled with \? quot;, whose values must be consistent with the constraints. Clearly there are (in nitely) many consistent ways to ll in the cells of table 1; one such way is shown in table 2.... ..."

Cited by 45

### Table 2. The mean and standard error of misclassi cation probability (MP) estimate evaluated under the contaminated (Mc, SEc) and uncontaminated (Mu, SEu) distributions for various estimators

"... In PAGE 11: ... The misclassi cation probabilities of these discriminant rules for each and combined groups evaluated under the contaminated and the uncontaminated distributions are obtained through simulations with test sample size of 2000 from each group. For brevity, only the means and the standard errors of the misclassi - cation probability estimates of the combined group are given in Table2 . We used 100 Monte Carlo samples in the study, so the standard errors of the mean mis- classi cation probability estimates are one tenth of the standard errors reported in Table 2.... ..."

### Table 2: Sensitization Probability Distribution

2003

"... In PAGE 5: ... It is evident from the results that the proposed scheme reduces the area overhead while providing high coverage in all the cases. In Table2 , we divide the interval [0, 1] into 8 equal subintervals and present the distribution of the num- ber of faults with a sensitization probability over these subintervals. In the table, an entry of x ! y under the interval [0:125; 0:25] indicates that the number of faults with a sensitization probability (= SPparity) in that interval went from x in the unprotected circuit to y (= SP ) in the circuit protected using the proposed scheme.... ..."

Cited by 9

### TABLE 1. KS tests of likelihoods: Probability that the \mirror quot; distribution and distribu- tions from di erent source catagories arise from the same underlying population.

### Table 3 Probability Function for AS in %

"... In PAGE 4: ... The computational advantage of AS can be observed in the sequences with slow uniform motion such as akiyo and grandma . Table3 gives the probability distribution of the various classifications under the adaptive search and Table 4 provides the failure rate of the classification methodology. It is evident from the results, those in sequences where the motion is slow, the classification probability pi,j(4MB) is the factor, which influences the computational complexity, and due to the reduced failure rate in these sequences, a high computational advantage is obtained.... ..."

### Table 15: Poverty and Inequality Group Weights Welfare Theil P0 P1 P2 P0*

"... In PAGE 47: ... The second (P0*) corresponds to the computation of the poverty rate under the standard assumption of a lognormal distribution of the within-group income, with endogenous mean and fixed variance. Table15 gives a static image of the differences between the two measures. At the aggregated level, P0* underestimates the poverty rate, but the results differ according to groups.... ..."

Cited by 1