cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
B1234
Level III

Relative Potancy 95% CI

Hi there, 

I am fitting a curve and applying a sigmoidal 4PL Rodbard model and testing parallelism. I have to determine the relative potency using the inflection points (IC50). So I calculate relative potency by dividing inflection point of the standard by the inflection point of the sample  (0.6479693 / 0.6477584 = 1.0003).

 

How would I create a 95% confidence interval for the relative potency?

 

 

Capture.PNG

 

 

 
 

 

4 REPLIES 4
Byron_JMP
Staff

Re: Relative Potancy 95% CI

Maybe try looking under the Relative Potencies outline bar, which should be below the graph under the Parallel Fit Parameter Estimates.  The relative potencies are calculated and the standard error is included.  (at least it is in JMP15). 

 

It may be that there isn't conventional method of calculating the confidence interval, so the JMP developers only included the standard error? In that case, if you happen to have JMP Pro, you can bootstrap the relative potency and use the bootstrap confidence limits.

JMP Systems Engineer, Health and Life Sciences (Pharma)

Re: Relative Potancy 95% CI

You can read the background information about ratio estimators here, to go with @Byron_JMP's answer.

Byron_JMP
Staff

Re: Relative Potancy 95% CI

The problem, even with the solution I proposed is that the interval around the relative potency is likely non-symmetric. So, a simple +/- the t-quantile*SE is going to overestimate on the low side and underestimate on the high side.

JMP Systems Engineer, Health and Life Sciences (Pharma)
B1234
Level III

Re: Relative Potancy 95% CI

Thank you @Byron_JMP and @Mark_Bailey ,

 

When I look at the relative potencies tab below the graph, it is not using the inflection points to generate the relative potency. For the bioassay I'm using I need to generate the RP using the relative potency. 

I understand what you are saying about the over/under estimation. I will read more on the link.

Capture.PNG