apascualgarcia
commented
4 years ago I created a new script, model_output_summaries.R that has creates a table containing results for a set of experiments selected. If all experiments are considered, a large table is created, and it can be used with different scripts model_output_summaries_plot$experiment.R to create plots with different combinations of results (an $experiment).
The issue remains open, just in case @jordan-klein wants to address the above points.
It is needed to code the outputs for the experiments D, E, F, G and H. Some of them should be split into sub-experiments, as it was made in experiment B. In the combined intervention experiment (H) note that we are dealing with a nested experiment, and there is no counterpart in the null-model, so the relevant comparison would be with respect to the simulation in which the added intervention is missing.
It is needed to include the new parameters for lock-down and self-distancing to retrieve the files from new experiments.
In lines 178 and 226 (at least) it is hard-coded the population size = 2000. This would be incorrect for experiments with different population sizes. Also note that it is not needed to compute the fractions because there are already files available with the fractions, so it is just needed to gather data from those.Edit: it is needed to compute the whole population fraction"
It is needed to compute a new variable measuring the number of simulations in which no deaths were retrieved (probability of outbreak). The number of realizations for each experiments can be inferred from the number of lines of the files, for non-testing simulations it is in general 500.