### Table 1: Estimation of the t-copula for the model from Example 4.1 with sample size n = 1000. The estimators of correlation and tail dependence perform equally well across the di erent marginals. The correlation estimates are accurate (low empirical standard deviation (std) and mean square errors (m.s.e.)). The estimators of the tail index of the copula and the tail dependence coe cents have high empirical variance and m.s.e.

"... In PAGE 23: ...rror (m.s.e.), as compared to the direct estimator (2.6). This results immediatelly through Algorithm 5.2 in improved estimates of the tail index (see Table1 for comparison). Then the implied tail dependence coe cients estimates (step (5) of the Algorithm 5.... ..."

### Table 8: Value-at-Risk and Expected Shortfall at the 95th Percentile. Normal vs. Student-t copula with DoF=12, 100K-path Monte Carlo simulation.

"... In PAGE 27: ... Let us now assume that each of the 100 reference names has an objective default intensity equal to 50 basis points, the remaining parameters unchanged. Table8 compares the two dependence assumptions in terms of the 95% Value-at-Risk and Expected Shortfall that they produce for a number of loss tranches. Where we de ne the Value-at-Risk, VaR := DL 1( ) and the conditional-VaR, CVaR := 1 1 R 1 DL 1(t)dt, where DL is the discounted loss.... ..."

### Table 2. Description of parameters used to calculate substrate dependent terms.

"... In PAGE 5: ...able 1. Description of parameters used in the substrate model shown in Figure 5. The turn to turn modeling approach consists of breaking up the inductor turns that lay on the substrate into two parts: top (ESL2) and bottom (ESL1). Table2 provides a description of the variables used in the equations to attribute substrate dependent characteristics to the model. Variable Description H (mm) Distance from turn to ground plane.... ..."

### Table 5: Expected Values of ATTNFED, Increasing Variables of Interest by

"... In PAGE 28: ...21). A sense of the substantive meaning of these results is demonstrated in Table5 . I calculated expected values of the dependent variab le for each variable of interest, holding the remaining terms in the model constant at their respective means.... In PAGE 28: ... Likewise, a one standard deviation increase in REALTBIL is expected to yield a 27 percent increase in congressional activity directed at monetary policy. -- Table5 about here -- Returning to Table 3, note that the controls for procedural norm s and rules in Congress, NEWCONG and COSPON , have large positive effects. Indeed, NEWCONG is by far the most important term in the model.... In PAGE 30: ... Note that the associations remain correctly signed and of similar size and significance levels to those in Model 1, suggesting that the basic model is robust. Yet judging from the expected values reported in Table5... In PAGE 31: ... While the overa ll results are not sensitive to this sampling of the data, the exchange rate coefficient estimate does not indicate a stronger effect than in the full sample. Expected values of the dependent variable ( Table5 ) reveal that there is basically no change in the role of the exchange rate as a predictor of congressional attentiveness, even when capital mobility is unambiguously high. However, the estimate of the credit channel effect, FINMIX,... ..."

### Table 1. Gaussian Copula Algorithm

"... In PAGE 6: ...Table 2. Gaussian Copula Algorithm in Matlab The Gaussian copula can be simulated using the algorithm in Table1 [Cherubini et al., 2004].... In PAGE 17: ...7 Expiry Time Variance JK JK2 Plain MC (d) Basket B3 Figure 8. Variance of the MC and IS algorithm as function of swap maturity Given the algorithm to simulate a T copula in Table 5, [Joshi and Kainth, 2004] remarks that at step 4 we are in the same situation of step 3 of the Gaussian copula simulation algorithm (see Table1 ). Indeed we can repeat the reasoning that led to the gaussian importance sampling algorithm.... ..."

### Table 3: Results for copula grammar

"... In PAGE 6: ... Table3 displays the results of the copula grammar. These results can be put in contrast with those obtained with a grammar that provides values that can be seen as baseline scores.... ..."

### Table 4: General Dependency

"... In PAGE 3: ... Thus, bugs in the design mechanisms responsible for maintaining atomicity (such as the bug demonstrated in Figure 1) will not be detected. Table4 shows that there may be dependencies between non- adjacent resources. In this example, the Store and Load are atomic.... In PAGE 4: ... A reference model that supports this MP generation scheme should therefore support service methods to initialize a resource (call this Initialize-Resource) to simulate an instruction to compute its effects on resource values (call this Simulate) and a method to read the current, possibly non-unique, values that may be held by a resource as a result of the previously simulated instructions (call this Read-Resource). For example when using the reference model to aid the gen- eration of Table4 the generator will use Initialize-Resource to ini- tialize R1, M[1000] and R3 with the required values, will call the Simulate method for the three instructions and finally call Read- Resource to produce the report of the final resource values. The call to Simulate for the Load instruction, for example, should com- pute that R2 could now hold either 0 or 1.... In PAGE 4: ... Predicting Results without Propagation The term non-uniqueness propagation refers to an instruction that uses an input resource with several allowed values and a target re- source, where the value or resource identity depends on the value of the non-unique resource. For example, in Table4 , the Load propa- gates the two allowed values from M[1000] into R2 and the Add propagates the non-uniqueness further into R3. It is relatively straight forward to implement the Simulate and Read-Resource method if it is assumed that the scenarios do not in- volve propagation.... In PAGE 4: ... This can be done according to the specific semantics of the instruction which is simulated. For example, assume that Simulate is simulating the Load in- struction of Table4 . The Values store entry for address 1000 should show that process 2 can view two values - 0 and 1.... In PAGE 5: ... Yet another limitation in the above approaches to propagation simulation occurs in the case of Store-Load true sharing collisions, where the Store must be simulated before the Load. Consider again the Store-Load collision of Table4 . The Store should already have been simulated by the reference model by the time the Load is sim- ulated, so that the Load simulation will be aware of the propagation done by the Load.... ..."

### Table 3: Estimation of the t-copula with di erent marginals model from Example 4.1, sample size n = 1000. We observe that the new method (5.3) for estimation of the tail dependence coe cients improves the empirical variance and mean square error (m.s.e.), as compared to the direct estimator (2.6). This results immediatelly through Algorithm 5.2 in improved estimates of the tail index (see Table 1 for comparison). Then the implied tail dependence coe cients estimates (step (5) of the Algorithm 5.2) have also quite satisfactory empirical variance and m.s.e.

"... In PAGE 23: ...lgorithms 3.3 and 5.2 are the same, we consider the di erences only with respect to the estimated tail dependence coe cients and tail index . In Table3 we summarize the results, see also Figure 6. We observe that the new tail dependence estimator has smaller emprical variance than (2.... ..."

### TABLE I The constants and variables of the thermophysical model. \Material Dependent quot; indicates that the term is dependent or based upon the object and material observed (including information about orientation, time, etc.), and \Measurable quot; indicates that the term can be accurately estimated for a typical scenario (given an accurate model hypothesis). The terms with a * can be measured (or approximated), but with questionable accuracy. Note that = l. \Time Dependent quot; indicates that the term is a function of time (or at least changes with time).

Cited by 1

### Table 3 Fit for dyadic dependence modelsa

2007

"... In PAGE 13: ...5)}. The first round, adding each of these terms or sets of terms, yields the model fits in Table3 . The standard Markov terms of triangle and individual star parameters again led to a non-converging model.... ..."

Cited by 6