The problem of detecting any differences between the distributions of two populations is addressed within the non parametric permutation framework of combined tests. Combined testing has been very useful to address the location, the scale, and the location/scale problems. The aim of the paper is to see whether combined testing is useful also for the general two-sample problem. The framework of combined testing for the general two-sample problem is presented and some tests are proposed. These tests are valid even when a non random sample of units is randomized into two groups. Type 1 error rate and power characteristics of the new tests are investigated and compared to former tests. It is shown that the new tests compare favorably with the former ones. An application to a very important socioeconomic problem is discussed.

Combination of distribution-free tests for the general two-sample problem with application to the social sciences

MAROZZI, Marco
2016-01-01

Abstract

The problem of detecting any differences between the distributions of two populations is addressed within the non parametric permutation framework of combined tests. Combined testing has been very useful to address the location, the scale, and the location/scale problems. The aim of the paper is to see whether combined testing is useful also for the general two-sample problem. The framework of combined testing for the general two-sample problem is presented and some tests are proposed. These tests are valid even when a non random sample of units is randomized into two groups. Type 1 error rate and power characteristics of the new tests are investigated and compared to former tests. It is shown that the new tests compare favorably with the former ones. An application to a very important socioeconomic problem is discussed.
File in questo prodotto:
File Dimensione Formato  
comm-stat-th-meth-2016.pdf

non disponibili

Descrizione: Articolo completo
Tipologia: Versione dell'editore
Licenza: Accesso chiuso-personale
Dimensione 217.16 kB
Formato Adobe PDF
217.16 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3664967
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact