Observation Depth Measure in a Sample in the Voivodeship Classification of the Property Market
Małgorzata Kobylińska
Department of Quantitative Methods, University of Warmia and Mazury in OlsztynRésumé
The problem of classification has long been an object of interest in many fields of knowledge. It allows homogeneous groups of objects to be obtained with respect to a given criterion. The selection of the appropriate distance measure, which is used in the clustering of multivariate objects, has an important effect on the obtained classification results. This paper uses observation depth measure in a sample ofvoivodeship classification, with respect to selected features concerning the property market in 2011. Voivodeships characterized by typical values for all analysed features were distinguished and those which could be considered outliers were so designated because of high or low values for the studied variables.
Mots-clés :
data classification method, outlying observation, observation depth measure in a sampleRéférences
Gatnar E., Walesiak M. 2004. Metody statystycznej analizy wielowymiarowej w badaniach marketingowych. Wydawnictwo Akademii Ekonomicznej im. Oskara Langego we Wrocławiu.
Grabiński T., Wydymus S., Zeliaś A. 1989. Metody taksonomii numerycznej w badaniach zjawisk społeczno- gospodarczych. PWN, Warsaw.
He X., Wang G. 1997. Convergence of Depth Contours for Multivariate Datasets. The Annals of Statistics, 25: 495-504.
Kobylińska M., Wagner W. 2010. Klasyfikacja obiektów wielocechowych z wykorzystaniem miary zanurzania obserwacji w próbie. Monografie i Opracowania 570. Oficyna Wydawnicza SGH in Warsaw.
Liu R.Y., Parelius J.M., Singh K. 1999. Multivariate Analysis by Data Depth: Descriptive Statistics, Graphics and Inference. The Annals of statistics, 27: 783-858.
Rousseeuw P.J., Ruts I. 1996. Bivariate Location Depth. Applied Statistics, 45: 516-526.
Rynek nieruchomości w Polsce w 2011 roku. No. 15. BRE Property Partner.
Tukey J. W. 1975. Exploratory Data Analysis. Reading, MA: Addison-Wesley.
Wasilewska E. 2009. Statystyka opisowa od podstaw. Podręcznik z zadaniami. Wydawnictwo SGGW, Warszawa.
Department of Quantitative Methods, University of Warmia and Mazury in Olsztyn
Licence
An Author declares that his paper has not been published before (under the same or another title, or is a part of another publication) and does not infringe copyrights of other persons**. At the same time, the Author transfers to the Publisher the exclusive right to publish and to circulate this work in print in the form of a non-serial journal publication and in a form of an electronic publication.
The journal is available on Creative Common license CC-BY-NC-ND
