### 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 9: The winner prompts the learner to demote Iambic. itself incorrect, treating it as correct (at least temporarily) can allow the learner to make progress. This is a variation on an idea used in statistical learning. A general class of algorithms that uses this sort of iterative approach to dealing with hidden structure is the class of expectation-maximization, or EM, algorithms (Dempster, Laird, and Rubin, 1977). EM deals with missing variable values (analogous to hidden structure) by using a guess at a model (analogous to a grammar) to estimate the missing values. The data, including the estimates for the missing values, are then used to select a new model, and the procedure is iterated. The procedure here proposed for Optimality Theory di ers from EM in that it is non-statistical, but shares the higher-level outline of using a model/grammar to estimate values for hidden structure, and then using those estimated values to select a new model/grammar.

1998

Cited by 5

### Table 3 Contrast and CNR measurements using AEC

2007

"... In PAGE 11: ...3.2 CNR The results of the contrast and CNR measurements are shown in Table3 and Figure 4. The CNR required to meet the minimum acceptable and achievable image quality standards at the 60 mm breast thickness have been calculated and are shown in Table 3 and Figure 4.... In PAGE 11: ....3.2 CNR The results of the contrast and CNR measurements are shown in Table 3 and Figure 4. The CNR required to meet the minimum acceptable and achievable image quality standards at the 60 mm breast thickness have been calculated and are shown in Table3 and Figure 4. The CNR required at each thickness to meet the limiting values for CNR in the European protocol are also shown.... ..."

### Table 2: Timing and causality.

1998

"... In PAGE 8: ... We propose here a framework which generalizes these notions (we follow the technique presented in [4]). In Table2 , we associate to each Hybrid primitive statement, its presence calculus and causality calculus. The pres- ence calculus involves only presences and boolean signals, it summarizes the constraints on presences and their relations to boolean signals.... In PAGE 8: .... The statement \on T quot; is optional. When it is omitted, then causality X gt; Y always hold. Table2 is commented now. For a signal X, the status of X a instant t shall be \absent quot; if t 62 TX, and the value of Xt otherwise.... In PAGE 10: ...= TY = T (1) k S = T n U (2) k U = true( pre(Y ) init 0 0 ) (3) k T gt; (X; Y; pre(Y )) (4) k U gt; U (5) k (T ; U) gt; S (6) k on T : pre(Y ) gt; U , T gt; U (7) k on U : U gt; Y , U gt; Y (8) k on S : (X; Y ) gt; dY=dX , S gt; dY=dX (9) Table 3: The example of table 1 : presence and causality calculus. This Hybrid program is obtained by applying the rules of Table2 until xpoint is reached. To apply these rules, we should have introduced the intermediate signal Z = f(X; Y ) ; this intermediate signal is simply denoted by dY=dX in the above program.... In PAGE 10: ... Selfexplanatory. Table2 is used as follows for executing a program. First, for each statement of the program, add to this program the associated presence and causality calculi, following Table 2.... In PAGE 11: ... This cannot be avoided in general, but is better performed most seldomly. Since causality constraints, as derived from Table2 , depend on the syn- tax of the program, we can try to rewrite this program di erently, with the objective of breaking most possible circuits [4]. Rewriting part of the pro- gram involving timing and booleans, can be performed using exactly the techniques of Signal language compilation, since the Signal clock calcu- lus has identical algebraic structure as our presences and boolean signals.... In PAGE 14: ...ignal at instant t, i.e., some preorder on the set fX1,: : : ,Xkg. 3. For X; Y signals of the considered hybrid system P , we de ne on T : X gt; Y (15) as X t Y holds 8t 2 T (16) which consequently requires T TX \ TY : (17) The rationale for infering causality constraints in Table2 is simple. Actions of computing are abstracted as term rewriting : the computation action y = f(x) is abstracted as \y can be substituted by f(x) quot;, which is encoded as the preorder x y, also written x gt; y.... ..."

Cited by 10

### Table 1: Four layer meta modeling structure. Order Description Example

2000

"... In PAGE 5: ...allInstances- gt;size lt; 10. The outlined meta modeling approach leads to the four layer structure in Table1 [2]. The object data row represents the data generated from a particular model, e.... ..."

Cited by 5

### Table 1. Values of for the copula of the bivariate t-distribution for various values of , the degrees of freedom, and , the correlation. Last row represents the Gaussian copula.

2002

Cited by 67

### Table 2. Cox proportional hazard models for progression to AIDS in 96 study participants

2005

"... In PAGE 5: ...eroconversion (range, 0.3-20.4 months). Most individuals were analyzed between 8 and 15 months after seroconversion (85 of 97) The few individuals who were analyzed within 6 months after seroconversion and later than 17 months after seroconversion were not different in total CD4H11001 T-cell numbers, viral load, activation markers, and HIV-specific CD4H11001 T-cell percentages. In univariate analyses, no significant predictive value of the number of cytokine-producing HIV-specific CD4H11001 T cells was found ( Table2 ; Figure 2). When adjusted for HIV viral load and CD4H11001 T cells, also no significant predictive effect of the numbers of single IFNH9253H11001 or single IL-2H11001 CD4H11001 T cells was observed.... ..."

### Table 1. Illustration of the inability of a flat correlation model to fit the market tranche quotes. The market quotes are as of the 30-Aug-05. The correlation in the Gaussian copula model was chosen via a least squares fit to the traded tranche premiums. Note that the [22-100%] tranche is not traded but can be implied from the rest of the capital structure and level of the index.

2005

"... In PAGE 3: ... Note that the [22-100%] tranche is not traded but can be implied from the rest of the capital structure and level of the index. The previous one factor Gaussian copula model leads to a semi-explicit pricing of CDO tranches but does not match the market prices (see Table1 ) and has therefore been extended in various directions. As in Hull and White [2004] or Kalemanova et al.... In PAGE 6: ... However, it obviously casts some doubt about the economic meaning of such a quantity and the ability to smoothly interpolate between tranches. 2 Stochastic correlation The inability of the Gaussian Copula Model to fit the market tranches, as illustrated in Table1 is well-known. Generally, we can say that it underprices the equity and senior tranches and overprices the mezzanine.... ..."

Cited by 2

### Table 8. Final Non-Plant Model Average Performance

### Table 1. Diagnostic data sets from the STANDARD Missile and the UCI Machine Learning Repository. All test sets corresponding to diagnostic problems. Nominal tests are analogous to test outcomes o(Tj) while continuous tests are analogous to the test measurements Tj. The Missing Values column indicates if data is missing for one or more of the attributes in the data set. The Ljubljana data sets correspond to data available for academic use only.

"... In PAGE 8: ... Since the purpose of this research is to explore the applicability of Bayesian models to fault diagnosis and prognosis under the UID program, we sought data from the US Navy to support this research but also used several data sets from the UCI Machine Learning Repository [20]. The specific databases are described in Table1 . Each of the UCI data sets, while medical in nature, reflects data collected on real diagnostic problems.... ..."