Exponential fit slopes comparisson

[Salvador Lluch]

Hi there, I hope you can help me. I have two exponential functions of the form y=a*b^x, and I need to evaluate if the slopes are significantly different. All tests I found out there are for linear models only and, as you may have guessed already, I´m not a Math person.

 

[katxt]

Which do you regard as the slope, a or b? Do you have data for both?
Have you tried using y=a.exp(mx) or ln(y) = A +mx and using the linear model test?

 

[Salvador Lluch]

I regard b as the slope. Transforming and then use the linear model test sounds like a good option, I didn´t want to do that because it will add methods to an already large paper, but it may actually be the easiest option, right?

 

[katxt]

I think so, unless your y=a*b^x analysis gave you SE and df for b. If it did, there is an easy conservative approach you can use.