Research Article
Random-Parameter Multivariate Negative Binomial Regression for Modeling Impacts of Contributing Factors on the Crash Frequency by Crash Types
Table 10
Zero-inflated multivariate Poisson 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 | 0.177 (1.41) | 0.515 (3.27) | −0.228 (−1.02) | 0.426 (4.48) | −0.0568 (−0.16) |
| | Traffic characteristics | | Interchange | 2 | —3 | — | −0.314 (−2.48) | — | −0.552 (−2.46) | | Black spots | 5 | 1.020 (16.68) | 0.859 (10.40) | 1.004 (7.92) | 1.050 (22.56) | 1.230 (5.18) | | AADT | 4 | 0.0000236 (10.04) | 0.00000909 (3.03) | 0.0000114 (2.62) | 0.0000282 (15.83) | — |
| | Speed characteristics | | 4 | −0.0134 (−5.96) | −0.00807 (−2.81) | −0.00999 (−2.21) | −0.0155 (−9.21) | — | | 4 | 0.0119 (5.27) | 0.00829 (2.85) | — | 0.0113 (6.76) | 0.0235 (2.48) | | 2 | 0.0209 (4.86) | — | — | 0.0228 (6.93) | — | | 4 | −0.0196 (−6.81) | −0.00939 (−2.54) | −0.0142 (−2.62) | −0.202 (−9.33) | — |
| | Geometric characteristics | | 2 | 0.000000154 (2.54) | — | — | 0.000000110 (2.45) | — | | 1 | — | — | — | — | 0.000239 (2.13) | | 2 | 0.00000041 (6.12) | — | — | 0.000000336 (6.75) | — | | 5 | 0.000567 (15.62) | 0.000460 (9.02) | 0.000606 (8.71) | 0.000595 (21.66) | 0.000735 (6.54) | | 1 | — | — | — | 0.000000093 (2.08) | — | | 1 | — | −0.000131 (−2.71) | — | — | — | | 3 | −0.000348 (−3.67) | — | −0.000545 (−3.07) | −0.000423 (−2.83) | — |
| | Sight characteristics | | 5 | −0.00644 (−4.78) | −0.00641 (−3.35) | −0.0106 (−3.51) | −0.00696 (−6.65) | −0.0137 (−2.61) | | 3 | −0.104 (−3.93) | −0.106 (−3.06) | — | −0.107 (−5.39) | — |
|
|
Total number of parameters: 53; log-likelihood: −16194.117; AIC: 32662.229; BIC: 33203.686; :41460.088. |
