Fig. 3

Summary of \(R_e\) inference on simulated data with time-varying delay distributions. A \(R_e\) estimates on simulated data, with observation delays gradually changing from a long (at time 0) to a short (at time 150) observation delay distribution. B \(R_e\) estimates on simulated data, with observation delays gradually changing from a short (at time 0) to a long (at time 150) observation delay distribution. A and B Each row corresponds to one of five \(R_e\) scenarios. Each column corresponds to a different delay distribution in the analysis. In the first two columns, delay distributions are fixed and either short or long. In the third column, delay distributions are allowed to vary when estimating(from short to long or long to short). In each plot, the ground truth \(R_e\) is shown as a black line, the median (over 100 replicates) estimate is shown as a green line and the 95% confidence interval is shown as a green ribbon