Detection of Anomalies in Water Networks by Functional Data Analysis

Table 2

Simulated data with amplitude, isolated, and shift outliers with = 0.05. Mean and standard deviation (in brackets) of the percentage of correctly () and falsely () identified outliers over 100 simulation runs.

	Amplitude		Isolated		Shift
Method
LRT	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
TRIM	74.0 (22.7)	1.3 (1.0)	90.2 (18.7)	1.1 (1.1)	94.8 (9.7)	0.9 (1.0)
POND	63.2 (27.2)	0.7 (0.9)	84.6 (24.8)	1.0 (1.0)	90.2 (19.2)	1.1 (1.0)
ISFE	16.2 (20.4)	2.6 (2.0)	100 (0)	2.9 (2.2)	63.2 (40.5)	3.0 (1.8)
RMAH	59.0 (24.7)	1.0 (1.2)	62.6 (28.4)	0.9 (1.0)	97.6 (7.7)	1.0 (1.1)
HDR	51.4 (20.2)	2.6 (1.1)	51.8 (19.9)	2.5 (1.0)	65.0 (21.0)	1.8 (1.1)
FB	23.2 (21.4)	0.1 (0.3)	28.2 (36.5)	0.07 (0.3)	18.8 (22.2)	0.1 (0.3)
OUG	0 (0)	5.0 (3.1)	44.4 (23.4)	3.6 (2.5)	94.8 (9.3)	3.4 (2.1)
FOM	24.4 (24.5)	0.2 (0.5)	14.0 (25.8)	0.3 (0.7)	10.0 (17.4)	0.2 (0.4)
FOADA	57.4 (27.7)	2.4 (1.6)	96.2 (17.9)	2.6 (1.8)	99.2 (6.3)	3.0 (1.9)