Comparing Means from Probability Distributions using aoctool and multcompare

Hello MATLAB community,

I have data from multiple samples that creates a normal distribution in semi-log space (logarithmic on the x-axis). I am interested in comparing the x-values for statistical differences in mean. I have multiple distributions to compare and am planning to use aoctool in combination with multcompare in order to utilize the Tukey multiple comparisons adjustment. Through the fit function, I have been able to extract the mean and standard deviation of these datasets.

The next step is where I am encountering the problem. The aoctool expects an input of the raw data and tries to fit a linear regression to it which is not what I am trying to do. I also need the output of the aoctool in order to utilize the multcompare function. What are your suggestions for getting around this and comparing the data? I have included an examples of my data below.

Thanks for your time

Answer 1

[Jeff Miller]

Maybe I am missing something but from your description it sounds like you could compare the mean log(x) values of the different tests with anova1 and then use multcompare.