### Table 2 presents the expected losses for all estimators as well as the (optimum) Bayes estimator and our suggested estimator S. Even though many methods in the table assume that the density g( ) is a Gamma density, our semi-parametric estimator which does not make such an assumption is superior to all estimated approaches. We re-emphasize here that the other methods are designed to obtain Bayes estimators, and they require considerable e ort for di erent Bayesian output, whereas our approach readily obtains, through a drawn sample of i, any desired posterior summary.

"... In PAGE 17: ... The data are shown in Table 2. Table2 . One way analysis of variance Group 1 Group 2 Group 3 Group 4 1 -2.... In PAGE 30: ... In the last three densities, h(t) is a moment generating function of any density. Table2 .... ..."

### Table 3.5: Nemesis test pattern generation bridge-fault statistics The three largest circuits are not included in Tables 3.4 and 3.3 because the formulas to be satis ed for these circuits with bridge faults are too large for our workstations. Two things can be observed from Tables 3.3 through 3.5. The rst is that this system detects many of the realistic bridging faults that are not detected by the SSA generated tests. The additional computation e ort results in a considerable increase in realistic fault coverage. One also notices that the ATPG times for the bridging faults is much greater than one would expect. Why this is is not yet understood.

1991

"... In PAGE 2: ... McCluskey and Buelow have shown a simple relationship between the defect coverage of a test set, the yield of the manufacturing process, and the quality level of the circuits that passed the test set assuming that defects are independently distributed on the chip [MB88]. Table3 from McCluskey and Buelow apos;s paper shows that to obtain the relatively modest quality level of 200 DPM in a circuit with a 90% yield requires that the test set detect 99.8% of the manufacturing defects.... ..."

Cited by 41

### Table 1 Computational e ort for the approximation of a M/D/1/K queuing system. The system can be obtained from Fig. 7(a) by removing the failure/repair subnet and by assuming that transition arrive is exponential. Let K = 50, =

1996

"... In PAGE 14: ... Let K = 50, = 1 10sec, and = 1 5sec. Table1 shows the computational e orts to achieve a certain approximation accuracy measured in terms of the relative error rel. It can be observed that our approach achieves better results with less computational e ort for the particular model under consideration.... ..."

Cited by 8

### Table 1 Computational e ort for the approximation of a M/D/1/K queuing system. The system can be obtained from Fig. 7(a) by removing the failure/repair subnet and by assuming that transition arrive is exponential. Let K = 50, =

"... In PAGE 14: ... Let K = 50, = 1 10sec, and = 1 5sec. Table1 shows the computational e orts to achieve a certain approximation accuracy measured in terms of the relative error rel. It can be observed that our approach achieves better results with less computational e ort for the particular model under consideration.... ..."

### Table 2: Participating sites

1997

"... In PAGE 6: ...embers of a single topology-d group, whose master was excalibur.usc.edu. Table2 lists the sites participating in this study and their location. Conducting such a large scale, widely distributed experiment required considerable e ort.... ..."

Cited by 39

### Table 1: E ort

1998

"... In PAGE 38: ... Because of this choice of x is was su cient to set m = 3 for n 27. Table1 shows the e ort we need to solve Problem (P) for 10 n 27. We use the abbreviations IT for the number of iterations, TT for the total... ..."

Cited by 2

### Table 2: E ort

1998

"... In PAGE 39: ...0 of the rst workstations with respect to the second ones. Table2 shows the e ort we needed to solve Problem (P) for 28 n 35, n = 38; 39. The meaning of the abbreviations is the same as before but with... ..."

Cited by 2

### Table 2: Number of System calls for Power Management System

"... In PAGE 11: ...Table 2: Number of System calls for Power Management System Table2 re ects the considerable e ort IBM spent on the power management system. Over 130 new calls were added in 11,437 lines of code.... ..."

### Table 2: Number of System calls for Power Management System

"... In PAGE 11: ...Table 2: Number of System calls for Power Management System Table2 re ects the considerable e ort IBM spent on the power management system. Over 130 new calls were added in 11,437 lines of code.... ..."