Research Article
Random-Parameter Multivariate Negative Binomial Regression for Modeling Impacts of Contributing Factors on the Crash Frequency by Crash Types
Table 4
Random-parameter multivariate negative binomial model estimation results.
| | Variables1 | No.2 | Rear-end estimate (t-stat) | Bumping-guardrail estimate (t-stat) | Other estimate (t-stat) | Noncasualty estimate (t-stat) | Casualty estimate (t-stat) |
| | Constant | 5 | 1.867 (1.77) | 15.393 (15.21) | 11.763 (0.02) | −0.604 (−3.49) | 0.867 (0.85) |
| | Traffic characteristics | | | | Interchange | 3 | —3 | — | −0.422 (−2.91) | −0.165 (−2.05) | −0.636 (−2.58) | | Black spots | 5 | 1.048 (11.25) | 0.859 (9.77) | 1.276 (7.17) | 1.052 (12.37) | 1.274 (4.97) | | AADT | 5 | 0.0000254 (7.90) | 0.00000936 (3.00) | 0.0000126 (2.52) | 0.0000322 (11.11) | −0.0000128 (−1.63) |
| | Speed characteristics | | 5 | −0.0120 (−3.37) | −0.00795 (−2.32) | −0.0161 (−3.10) | −0.0180 (−5.34) | −0.0165 (−1.89) | | 5 | 0.00999 (2.76) | 0.00788 (2.29) | 0.0120 (2.22) | 0.0103 (3.07) | 0.0231 (2.30) | | 2 | 0.0200 (2.92) | — | — | −0.0326 (4.93) | — | | 4 | −0.0189 (−4.02) | −0.00948 (−2.12) | −0.020 (−3.18) | −0.0243 (−5.44) | — |
| | Geometric characteristics | | 2 | 0.000000173 (1.98) | — | — | 0.00000017 (1.91) | — | | 1 | — | — | — | — | 0.000278 (2.44) | | 2 | 0.000000359 (3.72) | — | — | 0.000000365 (3.75) | — | | 5 | 0.000557 (9.86) | 0.000468 (8.12) | 0.000672 (7.32) | 0.000685 (11.07) | 0.000747 (6.44) | | 1 | — | −0.000133 (−2.48) | — | — | — | | 3 | −0.000319 (−2.34) | — | — | −0.000453 (−3.19) | −0.000608 (−1.85) | | 1 | — | — | — | −0.000420 (−2.17) | — |
| | Sight characteristics | | 5 | −0.00493 (−2.29) | −0.0059 (−2.61) | −0.0107 (−3.01) | −0.00706 (−3.44) | −0.0135 (−2.51) | | 2 | — | −0.100 (−2.40) | — | −0.0878 (−2.06) | — |
|
|
Total number of parameters: 56; log-likelihood: −14326.558; AIC: 28985.114; BIC: 29610.316; :23043.735. |
