"... In PAGE 21: ... Because these four models are non-nested rationalizations of the s ame data, some divergence exists in t he elasticity estimates across the four models, with those from t he log-infrequency of purch ase model the most plausible. The last row of Table9 gives the sample me an of the elasticity of postal services demand with respect to the probability of compute r own ership. For the log-infrequency model, this mean elasticity implies that increases in th e p robability computer owners hip bring about reductions in the demand for postal delivery services at the household-level.... ..."

"... In PAGE 7: ...1 Security Services The OSI model speci es fourteen security services [12]. (They are summarized in Table1 ). These services, however, are not uniformly applicable to all types of objects being restricted by policy:... ..."

"... In PAGE 17: ... (Capital is held fixed so that we can focus on resource shifts, without the additional complication of resource accumulation.) With the exception of substitution and scale elasticities, which are drawn from the literature (see Table1 ), model parameters are calibrated to social accounting data from the 1999 revision of the Global Trade Analysis Project (GTAP) version 4 data base. The GTAP dataset includes information on national and regional input-output structure, bilateral trade flows, final demand patterns, and government intervention, and is benchmarked to 1995.... ..."

"... In PAGE 8: ... w.r.t. predictor xpjt may be written as: elijt = ( ij ? MSjt)xpjt @Yjt @xpjt (16) Just like for the parametric MNL model cross-elasticities for all brands i 6 = j are equal. Substituting for @log(Ajt)=@xpjt the expression for ANNAM apos;s (cross-) elas- ticities is: elijt = ( ij ? MSj)xpjt(apj + Kj X k=1 bkjcpkjhkjt(1 ? hkjt)) (17) It subsumes the well-known equation for the parametric MNL model as special case: elijt = ( ij ? MSj)xpjtapj (18) Table2 contains price elasticities and price cross-elasticities for the average price of each brand (with prices of the other brands also set to their average values). Contrary to MNL ANNAM indicates lower absolute values both for elasticities and cross-elasticities.... ..."

"... In PAGE 4: ... Although the study of structure-activity relationships is not within the scope of this paper, we speculate that ligands of DH and AE contain a subset of compounds with particular features, such as steroids, which render them more easily distinguishable from the bulk of nonactives. Because, in this study, the sample size is not kept fixed but allowed to vary according to the classification, high enrichment factors can be achieved with a high number of false negatives if the sample size is small ( Table5 ). For example, TV, yielding high enrichment factors, predicts in many cases a smaller number of compounds to be active than do other methods (Table 5).... In PAGE 4: ... Because, in this study, the sample size is not kept fixed but allowed to vary according to the classification, high enrichment factors can be achieved with a high number of false negatives if the sample size is small (Table 5). For example, TV, yielding high enrichment factors, predicts in many cases a smaller number of compounds to be active than do other methods ( Table5 ). For this reason, recall values were also calculated, which give the percentage of all actives which have been retrieved (Figure 1b).... ..."

"... In PAGE 10: ... The following tables provide some descriptive statistics on the average characteristics and consumption pattern. Table1 : Average Usage Statistics Monthly Usage per Individual Average Minimum Maximum Voice (Minutes) 167.99 0 11845.... In PAGE 18: ... Results and Discussion We first estimate the model without demographics and other explanatory variables. The results of the structure parameters are presented in Table1 . The standard errors are reported in parenthesis.... ..."

"... In PAGE 4: ... e should equal zero, within experimental error, in all experiments (for the system will always return to its initial state oflowest energy), Ro should equal the number of imposed revolutions and T should be a constant, depending on the elastic modulus and the viscous resistance, that is independent of the number of imposed revolutions; t is time of unwinding. The values obtained from such fitting are presented in Table1 . As can be seen, e averages to zero over the three experiments and Ro approximately equals the number of initial winding revolutions in each case.... ..."