A Method to Identify Changes in Pedestrian and Cyclist Collision Rates
In this study actual collision counts were used to create a longitudinal confidence region around the minimum collision rate of an intersection. This procedure takes into account both the mean and the variance of the collision counts to estimate the minimum underlying collision rate. However, we have shown that the variance for an individual intersection may still be too large to identify any statistical change in risk. Therefore, intersections are subsequently clustered and confidence intervals are constructed for each cluster.
