Listy Biometryczne – Biometrical Letters vol. 39(2), 2002 pp.53-61
|
FORECAST OF THE EXPECTED NONEPIDEMIC MORBIDITY OF ACUTE DISEASES
Andrés M. Alonso1, Armando Aguirre2, Edilberto González3 1Department of Mathematic, Universidad Autónoma de Madrid, Facultad de Ciencias, CXV – Despacho 606, 28049 Madrid, Spain 2Unit of Experimental Clinic Research, Hospital La Candelaria de Tenerife, Spain 3Unit of Epidemiological Surveillance, Instituto de Medicina Tropical Pedro Kouri, Cuba |
In epidemiological surveillance it is important that any unusual increase of reported cases be detected as rapidly as possible. Reliable forecasts based on a suitable temporal model of an epidemiological indicator are necessary to estimate the expected nonepidemic indicator and elaborate an alert threshold. First, we present a method for identifying and replacing the abnormal values in the time series, then we apply the autoregressive integrated moving average approach to forecast the expected nonepidemic morbidity of acute respiratory infections and acute diarrhoeal diseases. Using this approarch, we are able to detect the starting data of new epidemic values under routine surveillance conditions.
epidemiological surveillance, time series, alert threshold.
