### Table 1. Standard binomial model.

"... In PAGE 5: ... The null hypothesis is defined as p H = q : . For all possible values of d, Table1 pres- ents the figures to compare our measure with the standard ones. To compute the Bayes Factor, we con-... ..."

### Table 3. Negative Binomial Model for PDO Crash Frequency

"... In PAGE 8: ... This can potentially give rise to model specification errors. Modeling Results Non-injury Crashes Before and During Work Zone Period Model 1 ( Table3 ) provides information on the factors that influence the frequency of police-reported crashes in the pre-work zone period. The Rho-squared value, which provides a measure of the model fit, indicates a reasonable fit.... ..."

### Table 6. Negative Binomial Model for KABC Crash Frequency

"... In PAGE 9: ... The exposure term is statistically non-significant. Model 4 ( Table6 ) shows the effects of the independent variables on injury-producing crashes in work zones. The goodness of fit statistic for the model is low implying that the explanatory variables are explaining relatively less variation in the data.... ..."

### Table 5. Negative Binomial Model for KABC Crash Frequency in

"... In PAGE 9: ... 0.13. This indicates that the effect of work zone length is largely unchanged relative to the before work zone period, implying that reducing work zone length is not a critical consideration in the reduction of adverse work zone impacts. Injury Crashes in Before and During Work Zone Period Model 3 ( Table5 ) presents the effects of the independent variables on injury-producing crashes in the before work zone period. Summary statistics for the model are reasonable and the mean ADT is statistically significant (5% level) indicating that increased mean ADT results in higher injury-producing crash frequency.... ..."

### Table 3: Estimates for global correlation in the Beta-Binomial model

2006

Cited by 4

### Table 2. Binomial model results for different number of components

### Table 2: Checkability of the binomial regression model

1998

"... In PAGE 8: ... The second stage assumes i i N ; 2 0 0 2 ; i = 1; :::; 20 with the illustrative informative priors, 2 IG(c; d) such that E( 2 ) = 10; V ar( 2 ) = 3; 2 IG(e; f) such that E( 2 ) = 1; V ar( 2 ) = 1 and N ??0 2 ; ?100 0 1 : Sampling based tting of this model is accomplished using Metropolis steps within a Gibbs sampler. Table2 summarizes the checkability of this model in terms of the I(d) and the interstage corre- lations using 1000 replications each providing 1000 pos- terior samples. We see that associations are weak, that d2j1 should be very e ective, d1 less so with the d2 apos;s o ering little promise.... ..."

Cited by 2

### Table 2: Class implementing the hierarchical Binomial Beta-Binomial model for the LOH data.

2001

Cited by 2

### Table 2: Class implementing the hierarchical Binomial Beta-Binomial model for the LOH data.

### Table 1 -- MLE Results for the Single Period Binomial Logit Model

1998

"... In PAGE 3: ... Maximum likelihood estimation ( MLE ) of the binomial logit model yielded parameter estimates and statistics for separation likelihood. (See Table1 .) Interpreting the signs and significance levels of parameter estimates is similar to linear regression.... ..."