Time Series Outlier Detection Based on Sliding Window Prediction

Table 3

Statistical analysis of both methods with optimal parameters of given dataset.

Parameters	LZ station				HYK station
	Water level		Daily flow		Water level		Daily flow
	One-sided; (6, 0.96)	Two-sided; (6, 0.95)	One-sided; (7, 0.96)	Two-sided; (6, 0.95)	One-sided; (6, 0.95)	Two-sided; (6, 0.96)	One-sided; (5, 0.96)	Two-sided; (6, 0.95)
TP	20	18	18	19	14	13	17	17
TN	704	704	706	704	713	710	710	708
FP	4	4	3	5	2	5	2	4
FN	2	4	3	2	1	2	1	1
Sensitivity	90.91%	81.82%	85.71%	90.48%	93.33%	86.67%	94.44%	94.44%
Specificity	99.44%	99.44%	99.58%	99.29%	99.72%	99.30%	99.72%	99.44%
PPV	83.33%	81.82%	85.71%	79.17%	87.50%	72.22%	89.47%	80.95%
NPV	99.72%	99.44%	99.58%	99.72%	99.86%	99.72%	99.86%	99.86%