Time series outlier detection: a new non parametric methodology (washer)
AbstractThe production and exploitation of statistical data for a large amount of high frequency time series must allow a timely use of data ensuring a minimum quality standard. This work provides a new outlier detection methodology (washer): efficient for timesaving elaboration and implementation procedures, adaptable for general assumptions and for needing very short time series, reliable and effective as involving robust non parametric test. Some simulations, a case study and a ready-to-use R-language function (washer.AV()) conclude the work.
V. BARTNETT, T. LEWIS, (1994), Outliers in Statistical Data, John Wiley & Sons, New York.
R.E. BENESTAD (2004), Record-values, non-stationarity tests and extreme value distributions, “Global
and Planetary Change”, vol. 44, issue 1-4, pp. 11-26.
M. DAHLBERG, E. JOHANSSON, (2000), An Examination of the Dynamic Behaviour of Local Governments
using GMM Bootstrapping Methods, “Journal of applied econometrics”, vol 5, pp.
L. KOVACS, D. VASS, A. VIDACS (2004), Improving Quality of Service Parameter Prediction with Preliminary
Outlier Detection and Elimination, IPS’2004, Budapest Hungary.
S. PAPADIMITRIOU, H. KITAWAGA, P.B. GIBBONS, C. FALOUTSAY, (2002), LOCI: Fast Outlier Detection
Using the Local Correlation Integral, Technical Report IRP-TR-02-09, Intel Research
L. SOLIANI (2005), Manuale di Statistica per la Ricerca e la Professione. Statistica Univariata e Bivariata,
Parametrica e Non-Parametrica per le Discipline Ambientali e Biologiche, Dipartimento di
Scienze Ambientali, Università di Parma.
P. SPRENT (1998) Data Driven Statistical Methods, Chapman and Hall, London.
P. SPRENT, N.C. SMEETON, (2001), Applied Nonparametric Statistical Methods (3rd ed.), Chapman
and Hall, London.