No matter how many they pursue, you can never say they will DEFINITELY
reach the market; you can only show the relationship between number of
'pursuits' and the probability of reaching a market.

If every compound has the same success rate, and all of these are
independent, then this is a binomial function -  typing functions into
e-mail is often not illuminating, but any statistics text should have
such a formula, and most software can calculate it for you

But, without context, it's impossible to say whether the assumptions
above are met.


HTH

Peter
>>> [EMAIL PROTECTED] 4/13/2004 9:45:52 AM >>>
I have a statistics-related question:

Success rate of a compound  reaching the market as a drug: 0.93%
Compound failure rate - not reaching the market: 99.07%

How may compounds a company should pursue so that with a 95%
confidence interval we can say that a company's compound will
definitely reach the market?

There are only two outcomes for a compound:  Either reaching the
market as a drug or failing in experimental studies and the project
being suspended.

Thanks in advance for your help.

Best regards,
Chandra
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)


.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to