東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

On the choice of the number of class...

Zhou, Wei.

Linked to FindBook

Google Book

Amazon

博客來

On the choice of the number of classes in the Chi-squared goodness-of-fit test.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	On the choice of the number of classes in the Chi-squared goodness-of-fit test./
Author:	Zhou, Wei.
Description:	104 p.
Notes:	Source: Masters Abstracts International, Volume: 46-04, page: 2147.
Contained By:	Masters Abstracts International46-04.
Subject:	Statistics. -
ISBN:	9780494360170

On the choice of the number of classes in the Chi-squared goodness-of-fit test.
Zhou, Wei.

On the choice of the number of classes in the Chi-squared goodness-of-fit test. - 104 p.

Source: Masters Abstracts International, Volume: 46-04, page: 2147.

Thesis (M.S.)--University of Manitoba (Canada), 2007.

Key words: chi-squared test, goodness-of-fit, Monte Carlo simulation.

ISBN: 9780494360170Subjects--Topical Terms:

517247
Statistics.

On the choice of the number of classes in the Chi-squared goodness-of-fit test.
LDR:01734nam 2200241 a 45 001 949561
005 20110525
008 110525s2007 ||||||||||||||||| ||eng d
020 $a 9780494360170
035 $a (UMI)AAIMR36017
035 $a AAIMR36017
040 $a UMI $c UMI
100 1 $a Zhou, Wei. $3 1064868
245 1 0 $a On the choice of the number of classes in the Chi-squared goodness-of-fit test.
300 $a 104 p.
500 $a Source: Masters Abstracts International, Volume: 46-04, page: 2147.
502 $a Thesis (M.S.)--University of Manitoba (Canada), 2007.
520 $a Key words: chi-squared test, goodness-of-fit, Monte Carlo simulation.
520 $a When applying Chi-squared goodness-of-fit test on a continuous distribution, one controversial question is how to group the data and then how many classes are used. In this thesis, we are investigating the most appropriate number of classes under the grouping method of "equiprobability intervals" by computer intensive Monte Carlo simulation. For continuous distribution with location and scale parameters, we examined the patterns of the power of test against the various numbers of classes when testing null hypothesis against a general alternative hypothesis in two cases, either on a specified distribution with known parameters, or on a family of distributions with unknown parameters. Finally we concluded the most appropriate number of classes as expressed in empirical formulae for the above two cases respectively. Even if a continuous distribution is not from the distribution family with location and scale parameters, the empirical formulae are applicable.
590 $a School code: 0303.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 $a University of Manitoba (Canada). $3 1018707
773 0 $t Masters Abstracts International $g 46-04.
790 $a 0303
791 $a M.S.
792 $a 2007