Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

Say I have two conditional media for bacteria growth: a bacteriocidal drug and control. I want to see the effect of my bacteriocide on culture growth, so I set up 6 flasks: three with drug, three without. I take three independent measurements from each flask at each of many timepoints. I model the growth of my bacteria with this equation:

$$ B(t) = \frac{B_{max}}{1+e^{k(t_{50}-t)}} $$

where $B(t)$ is the measured density of bacteria at a given time and $t_{50}$ is the time at which the bacteria are at half their max. I want to test differences in $B_{max}$ and $t_{50}$ for my drug vs. control, so I will use GEE.

How can I specify a custom, non-linear equation for GEE so I can get parameter estimates, confidence intervals, and significance values for my two treatments? Keep in mind that this is an example and I am looking for a way to do this with any model equation. The GEE dialogue only provides some basic ones, and I can't find a way to expand upon these. Alternatively, the nonlinear regression options don't appear to allow multiple levels of repeated measures.

share|improve this question
Is R an acceptable solution? –  Roman Luštrik Jan 27 '14 at 9:57

2 Answers 2

I think difference-in- difference methodology could work out in your case if you want to find the difference in what you mentioned between the control and treatment groups.

share|improve this answer
If I am reading various online resources correctly, difference-in-difference is just another term for multilevel regression. I know that I can do this with SPSS's Generalized Linear Regression procedure (or even a carefully planned non-linear regression in certain circumstances) but these aren't the same as GEE unless I am doing linear regression (which I am not). –  DocBuckets Nov 12 '12 at 21:00

Take a look at IBM Link for GEE in SPSS:


share|improve this answer
Welcome to CV, @user20293. One of our goals here is to create a permanent record of statistical information in the form of answers to questions. As such, one of the things we worry about is linkrot. In light of this, would you mind providing a summary of the info at the link in case it goes dead in the future? Since you're new here, you may want to read our about page & our FAQ, which cover topics like this about CV. –  gung Jan 30 '13 at 23:41

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.