error propagation involving log-transformed values

JFS

New Member
#1
I have a question about error propagation and log-transformed data.

I have a measured value of 'A' (=5.20) and an associated standard error (=0.29).

I have a measured value of 'B' (=10.07) and an associated standard error (=0.65).

I need to report the difference in log-transformed values of these two quantities:

log(10.07) - log(5.20)

= 1.003 - 0.716

= 0.287

How do I calculate the standard error for this value?

I would greatly appreciate any suggestions. Thank you very much.
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
I don't have a direct answer for your, but I tried to do this earlier in the year (but was with risks). I remember you needed to create a pooled SE, and can't remember how weighting was involved.