Consistently Estimating Link Speed Using Sparse GPS Data with Measured Errors

Data sources using new technology such as the Geographical Positioning System are increasingly available. In many different applications, it is important to predict the average speed on all the links in a network. The purpose of this study is to estimate the link speed in a network using sparse GPS data set. Average speed is consistently estimated using Indirect Inference approach. In the end, the Monte Carlo evidence is provided to show that the results are consistent with parameter estimates.