Box-plot of the distribution of epidemic sizes larger than 10%.

<p>Comparison of the distribution of epidemic sizes for SIR simulations performed on the contact sensors network (CSN), the surrogate contact network without zeros in the contact matrix of link densities and with weights randomly drawn from the distribution of contact durations registered by sensors (), and the same surrogate contact network but with weights randomly drawn from a negative binomial fit of the distribution of contact durations registered by sensors in several similar environments (. For each box, the central mark stands for the median, its edges represent the 25^th and 75^th percentiles. The whiskers extend to the most extreme data points not considered outliers, while outliers are plotted individually. Points are drawn as outliers if they are larger than <i>a</i> + <i>h</i>(<i>b</i> − <i>a</i>) or smaller than <i>a</i> − <i>h</i>(<i>b</i> − <i>a</i>), where <i>a</i> and <i>b</i> are the 25th and 75th percentiles, respectively and <i>h</i> is the maximum whisker length set by default to <i>h</i> = 1.5. (1000 simulations for each value of the ratio <i>β</i>/<i>μ</i>. <i>β</i>/<i>μ</i> ∈ {6, 8, 10, 12, 14, 16, 18, 20, 30, 40}).