The first section describes two algorithms used to detect outliers ( Filzmoser et al., 2008 Peña & Prieto, 2001) based on projection pursuit. However, the disadvantage of the projection pursuit methods is that there are high requirements in terms of computational load, which increments significantly when there is an increase in the variables considered. The projection pursuit methods easily identify atypical observations, and they have the advantage that it is not necessary to know the data distribution. In multivariate methods, there exist two approaches to identify outliers: those based on the distances of the observations in the data center and those projecting the original data. In this paper we discuss some multivariate methods for detecting outliers. However, a controversial question is: What should we do when the outliers derive from the intrinsic heterogeneity of the data?

In the case of errors in measurement or data entry to the base, it is relatively simple to correct them and it is convenient to eliminate the obvious mistakes. The detection of outliers depends on the type of error (or cause) in the data. Other times, the outliers need to be removed or downweighted different causes motivate different procedures. Sometimes the detection of outliers is the first step in statistical analysis. Intrinsic heterogeneity: the observed elements belong to the target population, but the inherent variability of the samples differs from the rest in their choices, attitudes or behavior. Sampling errors: the incorporation of sample statistical units from different populations to the target population. Intentional errors of response from the respondents. Measurement errors, collection or transcription. They could affect the normality condition that is necessary in many multivariate techniques. They could mask the data pattern and distort the results, so the conclusions would be completely different without their presence. In this respect, Uriel Jiménez and Aldás Manzano (2005) point out: It is important to detect these observations since they affect the data analysis in different ways.

Palavras-chave: Outliers, busca de projeções, curtose, empresas argentinas.ĭatabases often show outliers observations, which present a different behavior from the majority. Existem distintos métodos multivariados para detectar dados atípicos, tais como os métodos baseados na distância ou os métodos de busca de projeções. A detecção de dados atípicos é importante na análise estatística por seu efeito na distorção das medidas descritivas e nos estimadores dos parâmetros. A heterogeneidade significativa entre observações pode ser uma consequência da presença de dados atípicos. Palabras clave: Datos atípicos, búsqueda de proyecciones, curtosis, empresas argentinas.Įste trabalho compara os métodos propostos por Peña e Prieto (2001), e Filzmoser, Maronna e Werner (2008) para detectar dados atípicos em empresas argentinas que cotizam suas ações no Mercado de Valores. Existen distintos métodos multivariados para detectar datos atípicos, tales como los métodos basados en la distancia o los métodos de búsqueda de proyecciones. La detección de datos atípicos es importante en el análisis estadístico por su efecto en la distorsión de las medidas descriptivas y en los estimadores de los parámetros. La heterogeneidad significativa entre observaciones puede ser una consecuencia de la presencia de datos atípicos. Keywords: Outliers, projection pursuit, Kurtosis, Argentinian companies.Įn este trabajo se comparan los métodos propuestos por Peña y Prieto (2001) y Filzmoser, Maronna y Werner (2008) para detectar datos atípicos en empresas argentinas que cotizan sus acciones en el Mercado de Valores. There are different multivariate methods to detect outliers, such as distance-based methods and projection pursuit methods.

The detection of outliers is an important task for the statistical analysis since they distort descriptive measures and parameters estimators. A significant heterogeneity between observations can be a consequence of the presence of outliers. In this paper, we compare the methods proposed by Peña and Prieto (2001), and Filzmoser, Maronna, and Werner (2008) to detect outliers in a set of Argentine companies that quote their shares in the Stock Exchange.

Dfind outliers in high dimension