论文部分内容阅读
为了改进用于分析大量影响因素的交通事故模型,采用基于马尔可夫链蒙特卡罗法和吉布斯抽样的条件自回归负二项模型来拟合过度散布性(由负二项过程拟合)、未观察异质性和空间相关性(由条件自回归过程拟合).统计检验显示,由于具有更小的预测误差和更强的参数估计,条件自回归负二项模型优于条件自回归泊松模型、负二项模型、零膨胀泊松模型和零膨胀负二项模型.研究结果表明,交通事故率和死亡人数与车道数、曲线长度、车道年平均日交通量和降雨量成正比.最大限速和最近医院距离与交通事故次数成反比,而与死亡事故次数成正比,这可能是由于过高的速度会引发更严重的事故以及救援伤者时丧失较长时间.
To improve the traffic accident model for analyzing a large number of influencing factors, a negative automorphic negative binomial model based on Markov chain Monte Carlo method and Gibbs sampling was used to fit for the excessive scatter (fitted by the negative binomial process ), No heterogeneity and spatial correlation were observed (fitted by conditional autoregressive process). Statistical tests showed that the conditional autoregressive negative binomial model was superior to the conditional self-regressive model due to the smaller prediction error and stronger parameter estimation Regression Poisson model, negative binomial model, zero-inflated Poisson model and zero-expansion negative binomial model.The results show that the traffic accident rate and the number of deaths and the number of lanes, the length of the curve, the annual average traffic volume and rainfall The maximum speed limit and the nearest hospital distance are inversely proportional to the number of traffic accidents and are directly proportional to the number of fatal accidents, which may be due to the fact that excessive speed will lead to more serious accidents and loss of time to rescue the injured.