Log in
Register|
aquapurple ☆ Bangalore, 2012-10-01 15:02 (4609 d 03:12 ago) Posting: # 9287 Views: 3,900 |
|
|
Dear All, I have a doubt regarding the new guideline of FDA. The guideline says that 95% upper confidence bound should be less than or equal to zero. Is there any chance that the 95% upper confidence bound will be equal to zero? How many decimal places should we keep for the same? Edit: Category and subject line changed. [Helmut] — Regards, Nee… |
|
Helmut ★★★   Vienna, Austria, 2012-10-01 16:06 (4609 d 02:08 ago) @ aquapurple Posting: # 9288 Views: 3,283 |
|
|
Hi Nee! ❝ The guideline says that 95% upper confidence bound should be less than or equal to zero. Is there any chance that the 95% upper confidence bound will be equal to zero? Yes, although very, very small. ❝ How many decimal places should we keep for the same? Good question! I guess most people calculate the CL in full precision and set up their software in such a way that a verbatim statement is printed. At least this is what I do. On the other hand FDA's SAS code for unscaled ABE does not include a rounding statement – which is required according to the ![[image]](img/uploaded/IAsmall.png) 2003 guidance. Since inclusion of the 90% CI requires rounding to two decimal places (i.e., four significant digits if <100% and five significant digits if ≥100%) rounding of the upper 95% CL to five significant digits may be an option. Interested in other opinions.— Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
d_labes ★★★ Berlin, Germany, 2012-10-01 18:42 (4608 d 23:32 ago) @ Helmut Posting: # 9289 Views: 3,197 |
|
|
Dear Helmut! ❝ ❝ How many decimal places should we keep for the same? ❝ ❝ Good question! ❝ ... rounding of the upper 95% CL to five significant digits may be an option. Interested in other opinions. I bet: If you report upper 95% CL = -1.0459E-94 (5 significant digits  ) you will be asked in a deficiency letter to comment.— Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2012-10-01 18:55 (4608 d 23:20 ago) @ d_labes Posting: # 9290 Views: 3,183 |
|
|
Dear Detlew! ❝ I bet: If you report Good point. Do you expect the chance to get a deficiency letter after rounding to five decimal places ( upper 95% CL = 0.00000) to be lower? As Nee asked in his original post: ❝ Is there any chance that the 95% upper confidence bound will be equal to zero? — Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
Ing. Helmut Schütz