A statistical method to evaluate extreme events in earthquake precurso
ry signals of electrical nature is proposed. The occurrence probabilit
y of seismo-electrical anomalies is computed by means of a parametric
time series approach. In order to avoid the influence of external nois
e, a preliminary filtering of the electrical signals is performed. In
a second step an autoregressive model to describe the residual time se
ries is suitably identified and fitted to the data. Finally, parametri
c inference of extreme events is carried out on the basis of the selec
ted model. The performance of the technique has been evaluated by cons
idering geoelectrical time series recorded by an automatic station, lo
cated on the Southern Apennine chain (Italy). This region was interest
ed by the November 1980, Irpinia-Basilicata earthquake and it is one o
f the most active areas of the Mediterranean region. Finally, a possib
le correlation between electrical anomalies and seismic events occurre
d in the area is discussed.