Suppose I have performed 2 different experiments to measure the number of cell in a mice. In the first experiment, we measure the cell in odd day (day 5,7,9,11,13) and in the second, we measure the cell in even day (4,6,8,10,12,14). We use different mice in these two experiments, so the data would be look like this :

EXPERIMENT 1 (Even day)

d4 -> 170

d6 -> 6000

d8 -> 3700

d10 -> 2700

d12 -> 1700

d14 -> 533


d5 -> 7500

d7 -> 9000

d9 -> 3252

d11 -> 3400

d13 -> 750

My question is, is there any statistical tool to analyse this kind of data ? I mean, can I just fit these data together (combine the even and odd experiment) to some non-linear function f(t) (non-linear curve fitting problem) ? Or should I put into consideration about the different experiment ? Or is there any statistical analysis for this (to analyse the data from different experiment) ?



Maybe you should start by checking if Experiment1 and Experiment2 give consistent results (default hypothesis) or not. Since you have no underlying model (parametric distribution) about the number of cell measured in an experiment (is that right ?), you should use non-parametric hypothesis tests. There are tons of non-parametric tests, see e.g. the wikipedia list. Most of these tests are available in common statistical packages or python pkgs...

If these tests do not reject your hypothesis that Exp1 and Exp2 give consistent results, you might think about combining data.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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