서브메뉴
검색
상세정보
Multivariate Test Statistics and Their Approximations: Some Problems. Elliott, Ronald S., Barcikowski, Robert S [microform]
Multivariate Test Statistics and Their Approximations: Some Problems. Elliott, Ronald S., Barcikowski, Robert S [microform]
상세정보
- 자료유형
- 마이크로피시
- 언어부호
- 본문언어 - English
- 청구기호
- 서명/저자
- Multivariate Test Statistics and Their Approximations: Some Problems. : Elliott, Ronald S., Barcikowski, Robert S - [microform]
- 발행사항
- 형태사항
- 20; 1
- 총서명
- ERIC Reports
- 주기사항
- 20p.; Paper presented at the Annual Meeting of the American Educational Research Association (Atlanta, GA, April 12-16, 1993).
- 초록/해제
- 요약This Monte Carlo study examines whether, given various numbers of variables, treatments, and sample sizes, in a one-way multivariate analysis of variance, Type I error rates of the test approximations provided by the BMDP program, the Statistical Analysis System (SAS), and the Statistical Package for the Social Sciences (SPSS) for Roys largest root, Hotellings trace, Wilks likelihood ratio, and P. C. S. Pillais trace meet the stringent criterion of J. V. Bradley (1978) for robustness. For each of 85 conditions, the Type I error rate for each of the above statistical tests was estimated based on 100,000 random samples per situation. Results indicate that in multivariate analysis of variance studies with relatively small numbers of subjects of around 15 per treatment level, or fewer, the current probability levels reported by SAS and SPSS are conservative for the F approximations based on Pillais trace and liberal for the F approximations based on Hotellings trace. The BMDP 4V program does not report Pillais trace and reports accurate probability values for Hotellings trace. All of these programs report accurate values for the F approximations based on Wilks likelihood ratio and for Roys largest root (SPSS does not report Roys largest root). Recommendations are made for specific conditions. Two tables and five figures present analysis results. (SLD)
- 복제주기
- Microfiche. . Springfield, VA : ERIC Document Reproduction Service. . microfiches ; 11×15 cm.
- 일반주제명
- 키워드
- 기타저자
- 기타저자
MARC
008980919s1993 us b 000 0 eng d■0010000414724
■001PCUL00342227
■002ED359231
■00520020731194330
■007heuumu---buua
■008980919s1993 us b 000 0 eng d
■040 ▼apcul
■0410 ▼aEnglish
■090 ▼a370.78▼bE68
■24500▼aMultivariate Test Statistics and Their Approximations: Some Problems.▼cElliott, Ronald S., Barcikowski, Robert S▼h[microform]
■260 ▼aU.S.; Ohio▼cApr 93
■300 ▼a20; 1
■440 0▼aERIC Reports
■500 ▼a20p.; Paper presented at the Annual Meeting of the American Educational Research Association (Atlanta, GA, April 12-16, 1993).
■520 ▼aThis Monte Carlo study examines whether, given various numbers of variables, treatments, and sample sizes, in a one-way multivariate analysis of variance, Type I error rates of the test approximations provided by the BMDP program, the Statistical Analysis System (SAS), and the Statistical Package for the Social Sciences (SPSS) for Roys largest root, Hotellings trace, Wilks likelihood ratio, and P. C. S. Pillais trace meet the stringent criterion of J. V. Bradley (1978) for robustness. For each of 85 conditions, the Type I error rate for each of the above statistical tests was estimated based on 100,000 random samples per situation. Results indicate that in multivariate analysis of variance studies with relatively small numbers of subjects of around 15 per treatment level, or fewer, the current probability levels reported by SAS and SPSS are conservative for the F approximations based on Pillais trace and liberal for the F approximations based on Hotellings trace. The BMDP 4V program does not report Pillais trace and reports accurate probability values for Hotellings trace. All of these programs report accurate values for the F approximations based on Wilks likelihood ratio and for Roys largest root (SPSS does not report Roys largest root). Recommendations are made for specific conditions. Two tables and five figures present analysis results. (SLD)
■521 ▼aRes
■533 ▼aMicrofiche.▼bSpringfield, VA▼cERIC Document Reproduction Service.▼emicrofiches ; 11×15 cm.
■650 4▼xEducation
■653 ▼aAnalysis of Variance▼aComputer Simulation▼aEstimation (Mathematics)▼aMonte Carlo Methods▼aMultivariate Analysis▼aProbability▼aRobustness (Statistics)▼aSample SizeB▼aBiomedical Computer Programs▼aHotellings t▼aLikelihood Ratio Tests▼aPillais Trace▼aRoys Largest Root▼aStatistical Analysis System▼aStatistical Package for the Social Sciences▼aType I Errors
■7001 ▼aElliott, Ronald S.
■7001 ▼aBarcikowski, Robert S.
■999 ▼a142; 150
한글
ENG
日本
中文
Việt Nam