Fig 1.
Monthly incidence of highly pathogenic influenza cases in Zhejiang province from April 2009 to November 2013.
The vertical axis represents the monthly incidence of highly pathogenic influenza cases based on the provincial surveillance data. X-axis shows the corresponding time of each monthly surveillance data of highly pathogenic influenza.
Fig 2.
A QQ plot of the highly pathogenic influenza incidence data against standard exponential quantiles.
Notice that a concave departure from the straight line in the QQ plot (as in this plot) is an indication of a heavy-tailed distributions.
Fig 3.
The sample mean excess of the highly pathogenic influenza incidence data against threshold values.
The figure shows the empirical mean excess function has a positive slope above the threshold 9, which indicates the data follows the GPD with a positive shape parameter.
Fig 4.
A QQ plot of residuals from GPD fit to the highly pathogenic influenza incidence data over threshold 9.
The solid line corresponds to standard exponential quantiles. The approximate linear QQ plot indicates the residuals from GPD follow the standard exponential distribution.
Fig 5.
Panels a, b, c and d show the point estimation of monthly incidence of highly pathogenic influenza in Zhejiang province at different quantiles (0.917, 0.958, 0.972 and 0.983). Estimated tail is plotted as solid line, actual data in circles. Vertical dotted lines are estimated monthly incidence of highly pathogenic influenza in Zhejiang province at different quantiles (0.917, 0.958, 0.972 and 0.983) (middle dotted line), lower confidence level (left dotted line) and upper confidence level (right dotted line).
Table 1.
Estimation of the incidence of highly pathogenic influenza of Zhejiang province under diffident quantiles.