GM ★ India, 20181113 05:29 (841 d 07:46 ago) Posting: # 19591 Views: 2,656 

Hello All, Hope all are good...!!! What is the difference between Fieller's Confidence Interval and Locke's Confidence Interval for the ratio of means? OR Both are same. We wanted to find the Fieller's Confidence Interval for the ratio of means as mentioned in the below guideline Kindly suggest some reference/example related to the clinical endpoint studies. Thanks in advance. — Best Regards, GM 
d_labes ★★★ Berlin, Germany, 20181113 14:35 (840 d 22:40 ago) @ GM Posting: # 19597 Views: 2,242 

Dear GM, » What is the difference between Fieller's Confidence Interval and Locke's Confidence Interval for the ratio of means? » OR » Both are same. The latter. » We wanted to find the Fieller's Confidence Interval for the ratio of means as mentioned in the below guideline » ... Can't see any hint in the direction of Fieller's confidence interval in your posted text. » Kindly suggest some reference/example related to the clinical endpoint studies. The mathematical apparatus behind Fieller's Confidence Interval is descibed in Hauschke D., Kieser M., Diletti E., Burke M. "SAMPLE SIZE DETERMINATION FOR PROVING EQUIVALENCE BASED ON THE RATIO OF TWO MEANS FOR NORMALLY DISTRIBUTED DATA" Statist. Med. 18, 93105 (1999) Have also a look into Rpackage PowerTOST , function CI.RatioF() .— Regards, Detlew 
GM ★ India, 20181115 09:22 (839 d 03:52 ago) @ d_labes Posting: # 19602 Views: 2,231 

Dear Detlew, Thank you for the reply. » » We wanted to find the Fieller's Confidence Interval for the ratio of means as mentioned in the below guideline » » ... » » Can't see any hint in the direction of Fieller's confidence interval in your posted text. As per the Guideline, To establish bioequivalence for the primary endpoint, the 90% confidence interval of the test/reference ratio of the mean percent change from baseline to week 12 in the inflammatory (papules and pustules) lesion counts and in the noninflammatory (comedones) lesion counts should be contained within [0.80, 1.25], using the per protocol (PP) population. As this is untransformed data, we have to use Fieller's Confidence Interval for the ratio of means. This is the reason, I was asked this question. Kindly provide any other thoughts on this. — Best Regards, GM 
Ohlbe ★★★ France, 20181115 10:04 (839 d 03:11 ago) @ GM Posting: # 19626 Views: 2,185 

Dear GM, » As per the Guideline [...] Could you please provide a link to the guideline you are refering to ? It may help anybody willing to help to provide a relevant answer. — Regards Ohlbe 
GM ★ India, 20181115 10:37 (839 d 02:37 ago) @ Ohlbe Posting: # 19627 Views: 2,181 

Dear Ohlbe, » » As per the Guideline [...] » » Could you please provide a link to the guideline you are refering to ? It may help anybody willing to help to provide a relevant answer. Pls see the point no. 14 of this link — Best Regards, GM 