Kukla Degiskenlerin T Istatistigi ile Aykiri Gozlemler Tespit Edilemez

Arzdar Kiraci

Istanbul University Econometrics and Statistics e-Journal, 2011, vol. 15, issue 1, 1-14

Abstract: In the current literature, in order to be able to detect a single observation as an outlier observation, this observation is represented by a dummy variable and the dummy variable is checked for statistical significance. For an observation to be an outlier observation, the thesis of significant t-statistics of dummy variable is used. This paper proves using a theoretic proof for simple regression model that this thesis is wrong and refutes this thesis using a counterexample. The example derived for this paper illustrates that an outlier observation detected by robust regression methods cannot be detected by the t-statistics of dummy variable. In addition, the effect of adding a dummy variable to regression on important regression statistics is investigated.

Keywords: Robust Regression; t-statistics; dummy variable; outlier; refute the thesis; simple regression model; detection problem; example (search for similar items in EconPapers)
JEL-codes: C2 C3 C51 C52 (search for similar items in EconPapers)
Date: 2011
References: Add references at CitEc
Citations: Track citations by RSS feed

Downloads: (external link)
http://eidergisi.istanbul.edu.tr/sayi15/iueis15m1.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ist:ancoec:v:15:y:2011:i:1:p:1-14

Access Statistics for this article

Istanbul University Econometrics and Statistics e-Journal is currently edited by Kutluk Kagan Sumer

More articles in Istanbul University Econometrics and Statistics e-Journal from Department of Econometrics, Faculty of Economics, Istanbul University Contact information at EDIRC.
Bibliographic data for series maintained by Ertugrul YASAR ( this e-mail address is bad, please contact ).