Inference and control of the nosocomial transmission of methicillin-resistant Staphylococcus aureus

Pei, Sen; Morone, Flaviano; Liljeros, Fredrik; Makse, Hernán; Shaman, Jeffrey L

In each iteration of the IF, the standard deviation of each parameter is shrunk by a factor a 2 ð0; 1Þ (or equivalently, the variance is discounted by a factor of a 2 ). In practice, the discount factor a can range between 0.9 and 0.99 (Ionides et al., 2006) . If a is too small, the algorithm may 'quench' too fast and fail to find the MLE; if it is too close to 1, the algorithm may not converge in a reasonable time interval. We stop the IF algorithm once the estimates of the ensemble mean stabilize. The number of iterations required for this convergence was determined by inspecting the evolution of posterior parameter distributions, as in Figure 2A . Note that once the ensemble mean stabilizes, increasing the iteration time will not affect the MLE, although it can lead to a further narrowing of the ensemble distribution.