### Table 1 Overview of Manipulations and Diagnostics

"... In PAGE 11: ... Considered together, the pattern of results strongly suggests a common underlying construct of concept interrelatedness. Table1 presents an overview of the converging manipulations and indicators that are explored. ____________________________ Insert Table 1 about here ____________________________ The experiments will focus on three methods for experimentally manipulating concept interrelatedness, and two methods for measuring interrelatedness.... In PAGE 11: ... Table 1 presents an overview of the converging manipulations and indicators that are explored. ____________________________ Insert Table1 about here ____________________________ The experiments will focus on three methods for experimentally manipulating concept interrelatedness, and two methods for measuring interrelatedness. The notion that concepts vary systematically in their interrelatedness will be supported if the experimental manipulations and measures consistently cohere together in locating particular concepts on the continuum of interrelatedness.... In PAGE 33: ... For behavioral indicators, the framework links responding to caricatures accurately, being highly influenced by nondiagnostic features, and being highly influenced by inter-category similarities. The framework also links all of the experimental manipulations shown in Table1 . It is possible to explain particular experimental results without hypothesizing a continuum between isolated and interrelated concepts.... ..."

### Table 5: Comparison of various criteria for prediction performance in three types of the MOS estimators: ten input linear estimator, thirteen input linear estimator, and thirteen input neural network.

"... In PAGE 8: ... Figures 4 (a), (b), (c) show the relations between the average perceived MOS and the estimated MOS that are predicted by the above three methods, with the first order regression lines. Table5 shows a comparison of several diagnostic attributes that evaluate the performance of the three models. The correlations coefficients are largest using the non-linear network model with the new interaural features; the correlation coefficients between measured and predicted MOS are 0.... ..."

### Table 2: Brooks and Gelman Convergence Diagnostics. Model Prior Iterations MPSRF # parameters # times .975 Largest .975

2002

"... In PAGE 15: ...Table2 summarizes convergence analysis using the Brooks and Gelman diagnostic. With 2000 sampler iterations, each of the LL models using informative priors on precisions of random effects showed little evidence of convergence problems, with the .... In PAGE 15: ...ere huge (e.g. 75.0, 111.2, 139.9). The model with vague priors is omitted from all subsequent discussion because of sampler convergence failure. [Insert Table2 about here] Use of squared covariate values in linear models induces correlations between the predictor variables, which in turn causes correlation between their coefficients. High cross-correlations among parameters slow MCMC sampler convergence.... ..."

Cited by 2

### Table 1: Iteration counts for iterative solutions of FIT2P. Algo- rithm switched phase at step 17.

"... In PAGE 22: ... The results for FIT2P are tabulated in Table 1. Table1 : Iteration counts for iterative solutions of FIT2P. Algo- rithm switched phase at step 17.... In PAGE 23: ...Indeed, as shown in Table1 , the number of PCG iterations taken to solve the normal equations generally increases as the IPM converges to a solution. On the other hand, when the two-phase algorithm switches to the RAE system (which occurs at the 17th IPM step), the number of SQMR iterations taken to solve the preconditioned RAE system generally decreases as the IPM solution converges.... ..."

### Table 1: A sample run. The desired behavior is both spin on input , both go straight on input quot; quot;. After thirteen iterations, convergence is reached.

1993

Cited by 73

### TABLE 2b: Parameter estimates of ARFIMA models for log Ut

### TABLE 2c: Parameter estimates of ARFIMA models for ut

### TABLE 2d: Parameter estimates of ARFIMA models for u*t

### Table 3: Convergence diagnostics for the Epilepsy data example; Columns 1=BUGS strategy, 2=centered parameterization, 3=centered parameterization with blocking. The diagnostics are calculated using the set of Splus routines CODA (Best et al., 1996). Always the default values for various parameters relating to the diagnostics have been used.

"... In PAGE 11: ... 4.3 Seed Data Example We consider the seed germination data of Crowder (1978, Table3 ), as analyzed by Breslow and Clayton (1993). The probability of seed germination pi in the ith plate is modeled as logit[pi] = 0 + 1xi + 2zi + 12xizi + bi; i = 1; : : : ; 21; where xi and zi are seed and extraction indicators, respectively.... In PAGE 14: ... This may be explained by the fact that in the rst parameteri- zation all the parameters have non-standard complete conditional distributions while in the latter only bjk apos;s have non-standard distributions. In the hierarchical case, we can use block updating methods for the param- eters 0; Base; Trt; BT and Age: Table3 shows illustrative behavior of the Gibbs sampler diagnostics for the three versions, viz., BUGS strategy, hierarchically re-parameterized strategy and the re-parameterized strategy with blocking.... ..."