AP05 wrote:> This seems like it's easy but I can't seem to get it.
>
>
> My equation is (a^b)/c = d.
>
> I know that I can solve for b via
>
> log(d*c) / log(a) = b
>
>
> However, I don't really have d in the first place I have d/sum(d)
> values...no d's...just d/sum(d) values that all add to 1.0
>
> So, for example, if my data is
>
> a=4,9,12,3
> b=unknown
> c=12,18,13,20
> d/sum(d)=43,45,23,18
d/sum(d) should indeed add to one, hence I do not understand the data
above...
If you do not have d (but only d/sum(d)), then you simply cannot
estimate b in (a^b)/c = d. At least, the solution is not unique.
BTW: In what sense is this related to R(-help)?
Uwe Ligges
>
> I need an estimate of b. How can I do model and code this?
>
> Again, b is a power term of a and b^a are divided by c.
>
> Thank you,
> Jeff
>
>
>
>
> Checked by AVG Free Edition.
>
> PM
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.