### Table 1. Rules for constructing the Markov model from empirical action and observation uncertainties.

2003

Cited by 1

### Table 2. Correlation Statistics for Simulated Versus Observed O3 Empirical Orthogonal Functionsa

"... In PAGE 7: ... [22] We correlate the simulated and observed EOFs to evaluate how well the models capture the spatiotemporal patterns of O3 variability. Correlation statistics are given in Table2 ; a slope of unity accompanied by a high coefficient of determination (r2) signals a good match. The slope reveals whether the model exhibits any bias in its represen- tation of the underlying processes, whereas the r2 gives the amount of variance in the observationally derived EOFs that is captured by the model.... In PAGE 7: ... This approach avoids introducing the model bias from the simulated EOFs into this evaluation. The corresponding statistics are shown in Table2 . MAQSIP reproduces the variability in the Figure 3b.... In PAGE 8: ...Table2... In PAGE 8: ...nd their associated time series (PC). PC comparisons are for the modeled and observed data projected onto the EOFs derived from observations. The reduced major axis method is used to calculate slopes of regression lines (lines of organic correlation) [Hirsch and Gilroy, 1984]. Table2 summarizes the ability of the MAQSIP and GEOS-CHEM models to reproduce the three principal EOFs (east-west, midwest-northeast, and southeast) in the O3 observations over the eastern United States in summer 1995 averaged over the corresponding model grids. ACH 10 - 8... ..."

Cited by 3

### Table 6.1-1 Empirical models of sulfur removal for observations where removal equalled or exceeded 80%

1978

### Table 2 Empirical Genomewide P Values for Markers with Observed Nonparametric Peak LOD Scores 11.0

"... In PAGE 7: ... [2001] present a similar analysis for eczema). Table2 also illustrates the utility of calculating anal- ysis- and data-specific thresholds when evaluating ge- nomewide significance levels. For example, at threshold 1.... ..."

### Table 5: Summary of main empirical methods used in HCI research Method Observational studies Surveys Experimental studies

"... In PAGE 9: ... In addition, ethnographic observation, participatory design, and scenario-based design are also being streamlined (Schuler 1993). Table5 summarizes the empirical methods and their strengths and weaknesses. Table 5: Summary of main empirical methods used in HCI research Method Observational studies Surveys Experimental studies... ..."

### Table 1. Typical timings for the initial stages on a 2.0GHz Xeon processor. Most of the time is taking up by querying the index table, although it is an empirical observation that the majority of multi-tracks are small. Verification by correlation typically removes 30% of the putatives matches.

"... In PAGE 7: ... If after this approximate reg- istration the intensity at corresponding points in the neighbourhood differ by more than a threshold, or if the implied affine intensity change between the patches is outside a certain range, then the match can be rejected. Table1 shows the time taken by, and the number of matches involved in, this process. The outcome of the indexing and verification stages is a large collection of putative multi-tracks ; a multi-track is simply the set of matches resulting from a query to the in-... ..."

### Table 6.1-3 Empirical model of sulfur removal for all observations for which values of all parameters were available

1978

### Table 4: Observed Extreme Returns of the daily SP-500, 1990-1996, and the Probabil- ity of that Return as Predicted by the Normal GARCH, Student-t GARCH model, the Extreme Value Estimation Method, and the Empirical Distribution. Observed Probabilities

"... In PAGE 13: ... Note that this problem can be bypassed by using T day data to obtain T day ahead predictions as suggested in the RiskMetrics manual. In Table4 we show the six highest and lowest returns on the daily SP-500 index from 1990 to 1996, or 1771 observations. We used the normal GARCH and Student-t GARCH models to predict the conditional volatility, and show in the table the prob- ability of an outcome equal to or more extreme than the observed return, conditional on the predicted volatility for each observation.... ..."

### Table 1 Empirical Size of 5% Tests10 Serially i.i.d. Data S 200 Observations

"... In PAGE 14: ...= 1.93. The exponential distribution is quite asymmetric. Both of the latter two distributions are heavy-tailed S to the point where the variance does not exist for the symmetric stable Paretian distribution with this index value. 8 The results of these calculations are given in Table1 . We observe that the concerns about the small-sample validity of the tests S in particular, the BDS test S are justified, at least at this sample length.... ..."

Cited by 1

### Table 2. Example 4.1. (a) The types of the sites. (b) The time series of observed densities. (c) Yearly empirical habitat type distributions.

in A Study of Electrofishing Bias in Terms of Habitat and Abundance Using Information-Theoretic Tools

2002

"... In PAGE 7: ...le 2(a). Let our biological knowledge be P(H = poor) = 0.7 and P(H = good) = 0.3, and our data on observed densities as shown in Table2 (b). We can see that each year more sites are electrofished from, until at year y5 electrofishing occurs at all sites.... In PAGE 7: ... We can see that each year more sites are electrofished from, until at year y5 electrofishing occurs at all sites. The resulting yearly empirical habitat distributions are shown in Table2 (c). We can now calculate our bias in the choice of sites for each year using D(Pyi(H) bardbl P(H)).... In PAGE 9: ...able 3. A set of possible bias-minimal ways of choosing a given number of sites with the habitat types of Example 4.2. * signifies inclusion. % of total number of sites s1 s2 s3 s4 s5 s6 s7 s8 s9 s10 10 * 20 * * 30 * * * 40 * * * * 50 * * * * * 60 * * * * * * 70 * * * * * * * 80 * * * * * * * * 90 * * * * * * * * * 100 * * * * * * * * * * Table2 (a)). We have a different series of observations, how- ever.... ..."