Modeling the Effects of Lake-Effect Snow Related Weather Conditions on Daily Traffic Crashes: A Time Series Count Data Approach

The study develops a crash count model establishing the relationship between lake-effect snow (LES) and traffic crashes. The methodological approach uses Integer-valued Generalized Autoregressive Conditional Heteroscedastic (INGARCH) model. Negative Binomial INGARCH model outperformed Poisson INGARCH model by managing the overdispersion and autocorrelation issues. The model also captured the temporal correlation and allowed nonnegative covariate effects. Overall, the proposed method enables safety personnel to better understand the impact of LES on increased crashes.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01746233
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jun 23 2020 3:18PM