Forecasting ESKAPE infections through a time-varying auto-adaptive algorithm using laboratory-based surveillance data

Table 1 Monthly forecasting of “ESKAPE” infections: assessment parameters and performance of the time-varying auto-adaptive algorithm

Bacterial species	No. of recorded infections (years 2002– 2011)	Time MA (months) ^a	MAPE ^b	Accuracy rate (%)
E. faecium	1,142	3	9.22	82.14
S. aureus	4,332	4	5.15	90.36
K. pneumoniae	2,284	12	3.09	89.61
A. baumannii	4,106	12	3.18	84.93
P. aeruginosa	5,165	3	4.94	87.95
Enterobacter species	1,293	6	7.23	84.34

Abbreviations: MA, moving average; MAPE, mean absolute percentage error.
^aThe time MA was identified using the autocorrelation function graph, as detailed in Methods.
^bMAPE value was calculated based on observed values and fitted values from 2002 to 2011.

ISSN: 1471-2334