### Table 1. (ABSTRACT v REAL) amp; (INDIVIDUAL v

### Table 10: Real versus simulated data|mature individuals in samples Real data, 175 individuals Sp85 Su85 Sp86 Su86

1999

"... In PAGE 25: ... In the actual data, only 18% to 19% of the marked whales captured were highly marked, compared to 27% in the simulated data. In Table10 we compare the numbers of marked mature individuals captured at each sampling occasion in the actual data with the average numbers obtained in the simulated data. We observe a good agreement between the percent values for real and simulated data.... ..."

### Table 7: Real versus simulated data|number of recaptured individuals with capture history w case

1999

"... In PAGE 24: ... The larger data set contained 1677 records, 229 belonging to marked individuals, with 16 of 229 captured on more than one occasion. Table7 compares the real to simulated data with respect to the frequencies of recaptured individuals with a given capture history w. For the simulated data we took the average number of individuals (rounded to the nearest integer) with a given capture history, based on 100 simulated data sets where the samples... ..."

### Table 2 Recessions in Latin America and other regions (1970-1992)

1996

"... In PAGE 2: ...1 The highly volatile nature of Latin America is reflected in the fact that the region has more, and deeper, recessions than do most other areas of the world.2 Table2 provides evidence on the frequency, length, and depth of recessions in Latin America, and places the experience in an... ..."

Cited by 6

### Table 1. Results of two factors analysis of variance, for each experimental condition, between performances of individuals at generation 200 in the simulated and in the real environment. We also compared, for each of the three conditions, the performance in the real environment of the 500 individuals of generation 200 with performance of individuals of generation 219 (i.e. individual that was trained also in the real environment). The analysis revealed a statistically significant effect in all of the three cases (see Table 2). This means that in all cases, performances increase significantly by continuing the evolutionary process in the real environment.

1995

"... In PAGE 14: ... To study the statistical significance of the obtained results we made a two factors ANOVA to compare, for each of the three conditions, the performances of the 500 individuals of generation 200 in the simulated environment and in the real environment. The analysis revealed that statistically significant different performances are obtained in the first two conditions but not in the last one (see Table1 ). This means that performances of evolved individuals in the conservative noise condition do not significantly differ in the simulated and in the real environment.... ..."

Cited by 122

### Table 1. Results of two factors analysis of variance, for each experimental condition, between performances of individuals at generation 200 in the simulated and in the real environment. We also compared, for each of the three conditions, the performance in the real environment of the 500 individuals of generation 200 with performance of individuals of generation 219 (i.e. individual that was trained also in the real environment). The analysis revealed a statistically significant effect in all of the three cases (see Table 2). This means that in all cases, performances increase significantly by continuing the evolutionary process in the real environment.

1995

"... In PAGE 14: ... To study the statistical significance of the obtained results we made a two factors ANOVA to compare, for each of the three conditions, the performances of the 500 individuals of generation 200 in the simulated environment and in the real environment. The analysis revealed that statistically significant different performances are obtained in the first two conditions but not in the last one (see Table1 ). This means that performances of evolved individuals in the conservative noise condition do not significantly differ in the simulated and... ..."

Cited by 122

### Table 1. Results of one-way ANOVA, for each experimental condition, between performances of individuals at Generation 200 in the simulated and real environments. Degrees of Ratio of Fisher Significativity Remarks

1995

"... In PAGE 11: ... To study the statistical significance of the results obtained we performed one way ANOVA to compare the performances of the 500 individuals of Generation 200 in the simulated environment and in the real environment for each of the three conditions. The analysis revealed that statistically significant different performances are obtained in the first two conditions but not in the last one (see Table1 ). This means that performances of evolved individuals in the conservative-noise condition do not significantly differ in the simulated and real environments.... ..."

Cited by 122

### Table 4: Summary of execution time errors by program at the minimum and maximum levels of optimi- zation for the programs on Table 3. RMS represents the root mean square error. The individual real and predicted execution times are given in [Saav92b].

1995

"... In PAGE 17: ...ages inside parenthesis). The error is computed as the relative distance to the real execution time. If a program exhibits a significantly larger positive prediction error at the maximum optimization level than it does with no optimization, then it is probably the case that the error is the result of ignoring non-invariant optimizations. In Table4 we see several pro- grams for which this is true. An analysis of the source code shows that in these cases, optimizers are applying optimizations that are not invariant.... ..."

Cited by 13

### Table 9: Real versus simulated data|number of highly marked captured individuals Data set Sp85 Su85 Sp86 Su86

1999

"... In PAGE 25: ...nd 26 in the larger data set. So we cannot expect to obtain very reliable estimated values for N. There were also fewer highly marked whales than expected in the real data sets. Table9 compares the numbers in the real data sets with the numbers in the simulated data. In the actual data, only 18% to 19% of the marked whales captured were highly marked, compared to 27% in the simulated data.... In PAGE 26: ... However, convergence to the estimates given was slow. Consideration of the number of parameters in this model and the limitations of the actual data, as compared to simulated data ( Table9 ), leads to the conclusion that the heterogeneity model has too many parameters for the data to support. There were too few highly marked whales in the actual data.... ..."