cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Discussions

Solve problems, and share tips and tricks with other JMP users.
%3CLINGO-SUB%20id%3D%22lingo-sub-655487%22%20slang%3D%22fr-FR%22%20mode%3D%22NONE%22%3ETest%20d'%C3%A9quivalence%20de%20moyennes%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-655487%22%20slang%3D%22fr-FR%22%20mode%3D%22NONE%22%3E%3CP%3E%3CSPAN%3EBonjour%2C%3C%2FSPAN%3E%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3EMon%20sujet%20est%20surtout%20une%20demande%20sur%20un%20test%20statistique%20et%20non%20sur%20l'utilisation%20de%20JMP%2C%20bien%20que%20j'aurai%20une%20petite%20question%20%C3%A0%20ce%20sujet%20en%20fin%20de%20discussion.%3C%2FP%3E%3CP%3E%3CBR%20%2F%3E%3CSPAN%3EJe%20suis%20en%20pleine%20incompr%C3%A9hension%20sur%20l'utilisation%20de%20deux%20m%C3%A9thodes%20diff%C3%A9rentes%20du%20test%20d'%C3%A9quivalence%20de%20moyennes%20TOST%20%2F%20Schuirmann%20(je%20parle%20bien%20d'%C3%A9quivalence%20et%20non%20de%20comparaison%20de%20moyenne%20!).%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3EVoici%20les%20donn%C3%A9es%20%3A%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3EMoyenne%201%3A%20101.0%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3EMoyenne%202%3A%20101.9%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3EBiais%20inf%C3%A9rieur%20accept%C3%A9%20(B-)%20%3A%20-1%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3EBiais%20sup%C3%A9rieur%20accept%C3%A9%20(B%2B)%20%3A%20%2B2%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3E%CF%831%3A%200.9%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3E%CF%832%3A%201.5%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3En1%3A%206%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3En2%3A%206%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3E%CE%B1%3A%200.05%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3Etcritique%3A%201.81%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3EVarpool%3A%201.52%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3E%CF%83pool%3A%201.23%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3EVoici%20les%20deux%20m%C3%A9thodes%20utilis%C3%A9es%20pour%20%C3%A9valuer%20l'%C3%A9quivalence.%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3E1%C3%A8re%20m%C3%A9thode%3A%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ECalcul%20de%20l'intervalle%20de%20confiance%20%3A%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ECI%20%3D%20(X1%20-%20X2)%20%2B%2F-%20t(1-2%CE%B1%2Cn1%20%2B%20n2%20-2)*%CF%83pool*(1%2Fn1%20%2B%201%2Fn2)%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ECe%20qui%20donne%20%5B%200.17%20-%201.66%20%5D%2C%20l'intervalle%20%C3%A9tant%20compris%20entre%20les%20bornes%20d'acceptation%20de%20l'%C3%A9quivalence%20%5B%20-1%20et%202%20%5D%2C%20le%20test%20conclue%20%C3%A0%20l'%C3%A9quivalence%20des%20deux%20moyennes.%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3E2e%20m%C3%A9thode%20%3A%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ECalcul%20des%20statistiques%20TLower%20et%20TUpper.%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ETL%20%3D%20%5B(X1%20-%20X2)%20-%20B-%5D%20%2F%20%CF%83pool%20*%20Racine%20(1%2Fn1%20%2B%201%2Fn2).%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ETU%20%3D%20%5B(X1%20-%20X2)%20-%20B%2B%5D%20%2F%20%CF%83pool%20*%20Racine%20(1%2Fn1%20%2B%201%2Fn2).%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3EJe%20trouve%20pour%20TLower%20%3A%202.69%20et%20TUpper%20%3A%20-1.52.%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ET%20critique%20avec%20%CE%B1%3A%200.05%20et%20ddl%2010%20%3A%201.81.%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3EAvec%20les%20deux%20t%20valeurs%2C%20je%20ne%20retrouve%20pas%20l'%C3%A9quivalence%20(t%20valeur%20-1.52%20%26gt%3B%20-1.81).%3C%2FSPAN%3E%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3EJ'aimerais%20donc%20comprendre%20pourquoi%20cette%20diff%C3%A9rence%20%3F%20Sachant%20que%20l'approche%20par%20l'intervalle%20de%20confiance%20reste%20extr%C3%AAmement%20rare%20(voir%20lien%20%3A%26nbsp%3B%3CSPAN%3E%26nbsp%3B%3C%2FSPAN%3E%3CA%20href%3D%22https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC2548392%2Fpdf%2FJAMMC-25-123.pdf%22%20target%3D%22_blank%22%20rel%3D%22nofollow%20noopener%20noreferrer%22%3Ehttps%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Fart...MMC-25-123.pdf%3C%2FA%3E)%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3ESinon%20pour%20l'utilisation%20de%20JMP%20(version%2017.0%20dans%20mon%20cas)%2C%20lorsque%20je%20lance%20un%20test%20d'%C3%A9quivalence%2C%20JMP%20demande%20une%20valeur%20d'%C3%A9cart%20pour%20attester%20%C3%A0%20partir%20de%20quand%20la%20non%20%C3%A9quivalence%20est%20consid%C3%A9r%C3%A9e%2C%20or%20je%20ne%20peux%20pas%20avoir%20un%20biais%20d%C3%A9centr%C3%A9%20(mon%20exemple%20%3A%20-1%20%3B%20%2B2)%2C%20puisqu'une%20seule%20valeur%20doit%20%C3%AAtre%20rentr%C3%A9e.%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3EMerci%20d'avance%20pour%20votre%20retour%20sur%20ce%20sujet%20%5E%5E%20!%3C%2FSPAN%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3CSPAN%3ECordialement%20!%3C%2FSPAN%3E%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-LABS%20id%3D%22lingo-labs-655487%22%20slang%3D%22fr-FR%22%20mode%3D%22NONE%22%3E%3CLINGO-LABEL%3EBasic%20Data%20Analysis%20and%20Modeling%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3EQuality%20and%20Process%20Engineering%3C%2FLINGO-LABEL%3E%3C%2FLINGO-LABS%3E%3CLINGO-SUB%20id%3D%22lingo-sub-655625%22%20slang%3D%22en-US%22%20mode%3D%22CREATE%22%3ERe%3A%20Test%20d'%C3%A9quivalence%20de%20moyennes%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-655625%22%20slang%3D%22en-US%22%20mode%3D%22CREATE%22%3E%3CP%3EJe%20pense%20qu'il%20y%20a%20une%20erreur%20dans%20votre%20formule%20IC.%20Elle%20devrait%20%C3%AAtre%20%3A%20%3CSTRONG%3EIntervalle%20de%20confiance%3C%2FSTRONG%3E%20%3CSPAN%3E%3D%20(%3C%2FSPAN%3E%20%3CSPAN%3Ex%3C%2FSPAN%3E%20%3CSUB%3E1%3C%2FSUB%3E%20%3CSPAN%3E%E2%80%93%3C%2FSPAN%3E%20%3CSPAN%3Ex%3C%2FSPAN%3E%20%3CSUB%3E2%3C%2FSUB%3E%20%3CSPAN%3E)%20%2B%2F-%20t*%E2%88%9A((s%3C%2FSPAN%3E%20%3CSUB%3Ep%3C%2FSUB%3E%20%3CSUP%3E2%3C%2FSUP%3E%20%3CSPAN%3E%2Fn%3C%2FSPAN%3E%20%3CSUB%3E1%3C%2FSUB%3E%20%3CSPAN%3E)%20%2B%20(s%3C%2FSPAN%3E%20%3CSUB%3Ep%3C%2FSUB%3E%20%3CSUP%3E2%3C%2FSUP%3E%20%3CSPAN%3E%2Fn%3C%2FSPAN%3E%20%3CSUB%3E2%3C%2FSUB%3E%20%3CSPAN%3E))%3C%2FSPAN%3E%20o%C3%B9%20%3CSTRONG%3Es%20%3CSUB%3Ep%3C%2FSUB%3E%20%3CSUP%3E2%3C%2FSUP%3E%3C%2FSTRONG%3E%20%3CSPAN%3E%3D%20((n%3C%2FSPAN%3E%20%3CSUB%3E1%3C%2FSUB%3E%20%3CSPAN%3E-1)s%3C%2FSPAN%3E%20%3CSUB%3E1%3C%2FSUB%3E%20%3CSUP%3E2%3C%2FSUP%3E%20%3CSPAN%3E%2B%20(n%3C%2FSPAN%3E%20%3CSUB%3E2%3C%2FSUB%3E%20%3CSPAN%3E-1)s%3C%2FSPAN%3E%20%3CSUB%3E2%3C%2FSUB%3E%20%3CSUP%3E2%3C%2FSUP%3E%20%3CSPAN%3E)%20%2F%20(n%3C%2FSPAN%3E%20%3CSUB%3E1%3C%2FSUB%3E%20%3CSPAN%3E%2Bn%3C%2FSPAN%3E%20%3CSUB%3E2%3C%2FSUB%3E%20%3CSPAN%3E-2)%3C%2FSPAN%3E%20.%3C%2FP%3E%0A%3CP%3EJ'ai%20utilis%C3%A9%20ce%20script%20JMP%20avec%20la%20bonne%20formule%20%3A%3C%2FP%3E%0A%3CPRE%3E%3CCODE%20class%3D%22%20language-jsl%22%3ENames%20Default%20to%20Here(%201%20)%3B%0A%0AxBar1%20%3D%20101.0%3B%0AxBar2%20%3D%20101.9%3B%0A%0At%20%3D%20Students%20t%20Quantile(%201-0.05%2F2%2C%206%20%2B%206%20-%202%20)%3B%0A%0An1%20%3D%20n2%20%3D%206%3B%0As1%20%3D%200.9%3B%0As2%20%3D%201.5%3B%0A%0Asp%20%3D%20Sqrt(%20((n1-1)*(s1%5E2)%2B(n2-1)*(s2%5E2))%20%2F%20(n1%2Bn2-2)%20)%3B%0A%0Aci%20%3D%20(xBar1-xBar2)%20%2B%20%7B-1%2C1%7D*t%20*%20Sqrt(%20(sp%5E2%2Fn1)%2B(sp%5E2%2Fn2)%20)%3B%3C%2FCODE%3E%3C%2FPRE%3E%0A%3CP%3EJ'obtiens%20le%20r%C3%A9sultat%20%5B-2%2C49120941403157%2C%200%2C691209414031555%5D%2C%20qui%20d%C3%A9passe%20l'intervalle%20d'%C3%A9quivalence%20indiqu%C3%A9%20de%20%5B-1%2C2%5D.%3C%2FP%3E%0A%3CP%3EJMP%20utilise%20la%20deuxi%C3%A8me%20m%C3%A9thode%20que%20vous%20illustrez.%20JMP%20calcule%20%C3%A0%20partir%20des%20donn%C3%A9es%2C%20et%20non%20%C3%A0%20partir%20de%20statistiques%20d'%C3%A9chantillons%20r%C3%A9capitulatifs%2C%20je%20ne%20peux%20donc%20pas%20v%C3%A9rifier%20la%20conclusion%20par%20la%20deuxi%C3%A8me%20m%C3%A9thode.%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-SUB%20id%3D%22lingo-sub-655955%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3ERe%3A%20Test%20d'%C3%A9quivalence%20de%20moyenne%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-655955%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3E%3CP%3ESalut%3CA%20href%3D%22https%3A%2F%2Fcommunity.jmp.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F24055%22%20target%3D%22_blank%22%3E%20%40Arkawa%3C%2FA%3E%20%3A%20Vous%20avez%20fait%20des%20erreurs%20math%C3%A9matiques%20%3B%3C%2FP%3E%3CP%3ELes%20corrections%20sont%20%3A%20%3A%3C%2FP%3E%3CP%3EIC%20%3D%20(X1%20-%20X2)%20%2B%2F-%20t(1-2%CE%B1%2Cn1%20%2B%20n2%20-2)*%CF%83pool*sqrt(1%2Fn1%20%2B%201%2Fn2)%3C%2FP%3E%3CP%3E%3CSPAN%3ETL%20%3D%20%5B(X1%20-%20X2)%20-%20B-%5D%20%2F%20(%CF%83pool%20*%20Racine%20(1%2Fn1%20%2B%201%2Fn2)).%3C%2FSPAN%3E%3CBR%20%2F%3E%3CSPAN%3E%20TU%20%3D%20%5B(X1%20-%20X2)%20-%20B%2B%5D%20%2F%20(%CF%83pool%20*%20Racine%20(1%2Fn1%20%2B%201%2Fn2)).%3C%2FSPAN%3E%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3E%3CSPAN%3ETL%20et%20TU%20peuvent%20%C3%AAtre%20d%C3%A9riv%C3%A9s%20de%20la%20formule%20CL%2C%20de%20sorte%20que%20les%20deux%20m%C3%A9thodes%20sont%20%C3%A9quivalentes.%3C%2FSPAN%3E%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-SUB%20id%3D%22lingo-sub-655958%22%20slang%3D%22en-US%22%20mode%3D%22CREATE%22%3ERe%3A%20Test%20d'%C3%A9quivalence%20de%20moyennes%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-655958%22%20slang%3D%22en-US%22%20mode%3D%22CREATE%22%3E%3CP%3ESalut%20%3CA%20href%3D%22https%3A%2F%2Fcommunity.jmp.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F5358%22%20target%3D%22_blank%22%3E%40Mark_Bailey%3C%2FA%3E%20%3A%20Votre%20t%20doit%20%C3%AAtre%20ajust%C3%A9.%3C%2FP%3E%3CP%3EDevrait%20%C3%AAtre%3C%2FP%3E%3CPRE%20class%3D%22lia-code-sample%20language-markup%22%3E%3CCODE%3Et%20%3D%20Students%20t%20Quantile(%201-0.05%2C%206%20%2B%206%20-%202%20)%3B%3C%2FCODE%3E%3C%2FPRE%3E%3CP%3ECela%20est%20d%C3%BB%20au%20fait%20que%2C%20pour%20l%E2%80%99%C3%A9quivalence%20%C3%A0%20alpha%20%3D%200%2C05%2C%20l%E2%80%99intervalle%20de%20confiance%20de%2090%20%25%20IC%20(1-2%20alpha%20IC)%20est%20%C3%A9valu%C3%A9.%3C%2FP%3E%3CBR%20%2F%3E%3C%2FLINGO-BODY%3E%3CLINGO-SUB%20id%3D%22lingo-sub-655975%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3ERe%3A%20Test%20d'%C3%A9quivalence%20de%20moyenne%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-655975%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3E%3CP%3EBonjour%2C%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3E%3CA%20href%3D%22https%3A%2F%2Fcommunity.jmp.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F5358%22%20target%3D%22_blank%22%3E%40Mark_Bailey%3C%2FA%3E%20et%3CA%20href%3D%22https%3A%2F%2Fcommunity.jmp.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F7073%22%20target%3D%22_blank%22%3E%20%40MRB3855%3C%2FA%3E%20merci%20pour%20vos%20retours%20tr%C3%A8s%20instructifs%20!!%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3EJe%20viens%20de%20tester%20%C3%A0%20la%20main%20les%20deux%20m%C3%A9thodes%2C%20pour%20l'intervalle%20de%20confiance%20je%20trouve%20%3A%20-0.37%20-%202.21%20indiquant%20une%20absence%20d'%C3%A9quivalence.%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3EPour%20la%20statistique%20je%20trouve%20%3A%20TL%20%3D%202.69%20et%20TU%20%3D%20-1.52%20pour%20des%20Tcritique%20de%20-1.81%20%2F%20%2B%201.81%2C%20non%20%C3%A9quivalence%20prouv%C3%A9e%2C%20est-ce%20que%20quelqu'un%20peut%20me%20v%C3%A9rifier%20svp%20%3F%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3E%3A)%3C%2Fimg%3E%20merci%20!%3C%2FP%3E%3CP%3E%26nbsp%3B%3C%2FP%3E%3CP%3ECordialement%20!%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-SUB%20id%3D%22lingo-sub-656099%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3ERe%3A%20Test%20d'%C3%A9quivalence%20de%20moyenne%20(TOST%20%2F%20Schuirmann)%20-%20sujet%20statistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-656099%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3E%3CP%3EBonne%20prise!%20Script%20modifi%C3%A9%20et%20nouveau%20r%C3%A9sultat.%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22ci.PNG%22%20style%3D%22width%3A%20445px%3B%22%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%22%20image-alt%3D%22ci.PNG%22%20style%3D%22width%3A%20445px%3B%22%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%22%20image-alt%3D%22ci.PNG%22%20style%3D%22width%3A%20445px%3B%22%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%22%20image-alt%3D%22ci.PNG%22%20style%3D%22width%3A%20445px%3B%22%3E%3Cspan%20class%3D%22lia-inline-image-display-wrapper%22%20image-alt%3D%22ci.PNG%22%20style%3D%22width%3A%20445px%3B%22%3E%3Cimg%20src%3D%22https%3A%2F%2Fcommunity.jmp.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F54439iF7A42533315957C5%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22ci.PNG%22%20alt%3D%22ci.PNG%22%20%2F%3E%3C%2Fspan%3E%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%3C%2FLINGO-BODY%3E
Choose Language Hide Translation Bar
Arkawa
Level III

Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

Bonjour,

 

Mon sujet est surtout une demande sur un test statistique et non sur l'utilisation de JMP, bien que j'aurai une petite question à ce sujet en fin de discussion.


Je suis en pleine incompréhension sur l'utilisation de deux méthodes différentes du test d'équivalence de moyennes TOST / Schuirmann (je parle bien d'équivalence et non de comparaison de moyenne !).

Voici les données :
Moyenne 1: 101.0
Moyenne 2: 101.9
Biais inférieur accepté (B-) : -1
Biais supérieur accepté (B+) : +2
σ1: 0.9
σ2: 1.5
n1: 6
n2: 6
α: 0.05
tcritique: 1.81
Varpool: 1.52
σpool: 1.23

Voici les deux méthodes utilisées pour évaluer l'équivalence.

1ère méthode:

Calcul de l'intervalle de confiance :

CI = (X1 - X2) +/- t(1-2α,n1 + n2 -2)*σpool*(1/n1 + 1/n2)

Ce qui donne [ 0.17 - 1.66 ], l'intervalle étant compris entre les bornes d'acceptation de l'équivalence [ -1 et 2 ], le test conclue à l'équivalence des deux moyennes.

2e méthode :

Calcul des statistiques TLower et TUpper.

TL = [(X1 - X2) - B-] / σpool * Racine (1/n1 + 1/n2).

TU = [(X1 - X2) - B+] / σpool * Racine (1/n1 + 1/n2).

Je trouve pour TLower : 2.69 et TUpper : -1.52.

T critique avec α: 0.05 et ddl 10 : 1.81.

Avec les deux t valeurs, je ne retrouve pas l'équivalence (t valeur -1.52 > -1.81).

 

J'aimerais donc comprendre pourquoi cette différence ? Sachant que l'approche par l'intervalle de confiance reste extrêmement rare (voir lien :  https://www.ncbi.nlm.nih.gov/pmc/art...MMC-25-123.pdf)

 

Sinon pour l'utilisation de JMP (version 17.0 dans mon cas), lorsque je lance un test d'équivalence, JMP demande une valeur d'écart pour attester à partir de quand la non équivalence est considérée, or je ne peux pas avoir un biais décentré (mon exemple : -1 ; +2), puisqu'une seule valeur doit être rentrée.

Merci d'avance pour votre retour sur ce sujet ^^ !

Cordialement !

5 REPLIES 5

Re: Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

I think there is an error in your CI formula. It should be: Confidence interval = (x1x2) +/- t*√((sp2/n1) + (sp2/n2)) where sp2 = ((n1-1)s12 + (n2-1)s22) / (n1+n2-2).

I used this JMP script with the correct formula:

Names Default to Here( 1 );

xBar1 = 101.0;
xBar2 = 101.9;

t = Students t Quantile( 1-0.05/2, 6 + 6 - 2 );

n1 = n2 = 6;
s1 = 0.9;
s2 = 1.5;

sp = Sqrt( ((n1-1)*(s1^2)+(n2-1)*(s2^2)) / (n1+n2-2) );

ci = (xBar1-xBar2) + {-1,1}*t * Sqrt( (sp^2/n1)+(sp^2/n2) );

I get the result [-2.49120941403157, 0.691209414031555], which exceeds the stated equivalence interval of [-1,2].

JMP uses the second method you illustrate. JMP calculates from the data, not summary sample statistics, so I can't verify the conclusion by the second method.

MRB3855
Super User

Re: Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

Hi @Mark_Bailey : Your t needs to be adjusted.

Should be 

t = Students t Quantile( 1-0.05, 6 + 6 - 2 );

This is because, for equivalence at alpha=0.05,the 90% CI (1-2alpha CI)  confidence interval is evaluated. 

 

Re: Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

Good catch! Amended script and new result.

ci.PNG

MRB3855
Super User

Re: Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

Hi @Arkawa : You've made some mathematical errors;

Corrections are: : 

CI = (X1 - X2) +/- t(1-2α,n1 + n2 -2)*σpool*sqrt(1/n1 + 1/n2)

TL = [(X1 - X2) - B-] / (σpool * Root (1/n1 + 1/n2)).
TU = [(X1 - X2) - B+] / (σpool * Root (1/n1 + 1/n2)).

 

TL and TU can be derived from the CL formula, so the two methods are equivalent.

Arkawa
Level III

Re: Test d'équivalence de moyennes (TOST / Schuirmann) - sujet statistique

Bonjour,

 

@Mark_Bailey  et @MRB3855 merci pour vos retours très insctructifs !!

 

Je viens de tester à la main les deux méthodes, pour l'intervalle de confiance je trouve : -0.37 - 2.21 indiquant une absence d'équivalence.

 

Pour la statistique je trouve : TL = 2.69 et TU = -1.52 pour des Tcritique de -1.81 / + 1.81, non équivalence prouvée, est-ce que quelqu'un peut me vérifier svp ?

 

:) merci !

 

Cordialement !

Recommended Articles