Web projektu DAR

Research Centre
Data - Algorithms - Decision Making

Established in 2005 under support of MŠMT ČR (project 1M0572)

Publications

Goodness-of-fit Testing Based on Hypothetical and Empirical Quantizations.

Typ:

Internal print

Authors:

Vajda I., van der Meulen E.C.

Edition name / line:

Interní publikace DAR -

Volume:

ÚTIA 2006/14

Publisher:

ÚTIA AV ČR

Serie:

Praha

Year:

2006

Number of pages:

Contact:

Ing. Igor Vajda, DrSc. (ÚTIA - Oddělení stochastické informatiky)

Keywords:

Goodness-of-fit, Spacings, phi-divergence, power divergence

attachment1:

Goodness-of-fit Testing Based on Hypothetical and Empirical Quantizations. [file]

Anotation:

Revised and updated version of the Research Report UTIA no. 2160. Pearson-type statistics are demonstrated to be special cases of pfi-divergences of hypothetical and empirical distributions quantized by hypothetical quantiles. New class of spacing statistics is introduced by means of empirical quantiles. General asymptotic theory of the new statistics is presented and applied to their subclass generated by the power divergences. Concrete examples and calculated values are presented in the form of tables.